kenaz 0.0.0-stage → 0.3.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/bin/kenaz.js +3 -0
- package/core/.agents/skills/remotion-best-practices/SKILL.md +61 -0
- package/core/.agents/skills/remotion-best-practices/rules/3d.md +86 -0
- package/core/.agents/skills/remotion-best-practices/rules/animations.md +27 -0
- package/core/.agents/skills/remotion-best-practices/rules/assets/charts-bar-chart.tsx +173 -0
- package/core/.agents/skills/remotion-best-practices/rules/assets/text-animations-typewriter.tsx +100 -0
- package/core/.agents/skills/remotion-best-practices/rules/assets/text-animations-word-highlight.tsx +103 -0
- package/core/.agents/skills/remotion-best-practices/rules/assets.md +78 -0
- package/core/.agents/skills/remotion-best-practices/rules/audio-visualization.md +198 -0
- package/core/.agents/skills/remotion-best-practices/rules/audio.md +169 -0
- package/core/.agents/skills/remotion-best-practices/rules/calculate-metadata.md +134 -0
- package/core/.agents/skills/remotion-best-practices/rules/can-decode.md +81 -0
- package/core/.agents/skills/remotion-best-practices/rules/charts.md +120 -0
- package/core/.agents/skills/remotion-best-practices/rules/compositions.md +154 -0
- package/core/.agents/skills/remotion-best-practices/rules/display-captions.md +184 -0
- package/core/.agents/skills/remotion-best-practices/rules/extract-frames.md +229 -0
- package/core/.agents/skills/remotion-best-practices/rules/ffmpeg.md +38 -0
- package/core/.agents/skills/remotion-best-practices/rules/fonts.md +152 -0
- package/core/.agents/skills/remotion-best-practices/rules/get-audio-duration.md +58 -0
- package/core/.agents/skills/remotion-best-practices/rules/get-video-dimensions.md +68 -0
- package/core/.agents/skills/remotion-best-practices/rules/get-video-duration.md +60 -0
- package/core/.agents/skills/remotion-best-practices/rules/gifs.md +141 -0
- package/core/.agents/skills/remotion-best-practices/rules/images.md +134 -0
- package/core/.agents/skills/remotion-best-practices/rules/import-srt-captions.md +69 -0
- package/core/.agents/skills/remotion-best-practices/rules/light-leaks.md +73 -0
- package/core/.agents/skills/remotion-best-practices/rules/lottie.md +70 -0
- package/core/.agents/skills/remotion-best-practices/rules/maps.md +412 -0
- package/core/.agents/skills/remotion-best-practices/rules/measuring-dom-nodes.md +34 -0
- package/core/.agents/skills/remotion-best-practices/rules/measuring-text.md +140 -0
- package/core/.agents/skills/remotion-best-practices/rules/parameters.md +109 -0
- package/core/.agents/skills/remotion-best-practices/rules/sequencing.md +118 -0
- package/core/.agents/skills/remotion-best-practices/rules/sfx.md +30 -0
- package/core/.agents/skills/remotion-best-practices/rules/subtitles.md +36 -0
- package/core/.agents/skills/remotion-best-practices/rules/tailwind.md +11 -0
- package/core/.agents/skills/remotion-best-practices/rules/text-animations.md +20 -0
- package/core/.agents/skills/remotion-best-practices/rules/timing.md +179 -0
- package/core/.agents/skills/remotion-best-practices/rules/transcribe-captions.md +70 -0
- package/core/.agents/skills/remotion-best-practices/rules/transitions.md +197 -0
- package/core/.agents/skills/remotion-best-practices/rules/transparent-videos.md +106 -0
- package/core/.agents/skills/remotion-best-practices/rules/trimming.md +51 -0
- package/core/.agents/skills/remotion-best-practices/rules/videos.md +171 -0
- package/core/.agents/skills/remotion-best-practices/rules/voiceover.md +99 -0
- package/core/.agents/skills/supabase-postgres-best-practices/AGENTS.md +90 -0
- package/core/.agents/skills/supabase-postgres-best-practices/CLAUDE.md +1 -0
- package/core/.agents/skills/supabase-postgres-best-practices/SKILL.md +64 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/advanced-full-text-search.md +55 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/advanced-jsonb-indexing.md +49 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/conn-idle-timeout.md +46 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/conn-limits.md +44 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/conn-pooling.md +41 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/conn-prepared-statements.md +46 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/data-batch-inserts.md +54 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/data-n-plus-one.md +53 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/data-pagination.md +50 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/data-upsert.md +50 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/lock-advisory.md +56 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/lock-deadlock-prevention.md +68 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/lock-short-transactions.md +50 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/lock-skip-locked.md +54 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/monitor-explain-analyze.md +45 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/monitor-pg-stat-statements.md +55 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/monitor-vacuum-analyze.md +55 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/query-composite-indexes.md +44 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/query-covering-indexes.md +40 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/query-index-types.md +45 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/query-missing-indexes.md +43 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/query-partial-indexes.md +45 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/schema-data-types.md +46 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/schema-foreign-key-indexes.md +59 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/schema-lowercase-identifiers.md +55 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/schema-partitioning.md +55 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/schema-primary-keys.md +61 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/security-privileges.md +54 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/security-rls-basics.md +50 -0
- package/core/.agents/skills/supabase-postgres-best-practices/references/security-rls-performance.md +57 -0
- package/core/.agents/skills/typesafe-ai/LICENSE +21 -0
- package/core/.agents/skills/typesafe-ai/SKILL.md +149 -0
- package/core/.agents/skills/vercel-react-best-practices/AGENTS.md +2934 -0
- package/core/.agents/skills/vercel-react-best-practices/SKILL.md +136 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/advanced-event-handler-refs.md +55 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/advanced-init-once.md +42 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/advanced-use-latest.md +39 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/async-api-routes.md +38 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/async-defer-await.md +80 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/async-dependencies.md +51 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/async-parallel.md +28 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/async-suspense-boundaries.md +99 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/bundle-barrel-imports.md +59 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/bundle-conditional.md +31 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/bundle-defer-third-party.md +49 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/bundle-dynamic-imports.md +35 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/bundle-preload.md +50 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/client-event-listeners.md +74 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/client-localstorage-schema.md +71 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/client-passive-event-listeners.md +48 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/client-swr-dedup.md +56 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-batch-dom-css.md +107 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-cache-function-results.md +80 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-cache-property-access.md +28 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-cache-storage.md +70 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-combine-iterations.md +32 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-early-exit.md +50 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-hoist-regexp.md +45 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-index-maps.md +37 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-length-check-first.md +49 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-min-max-loop.md +82 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-set-map-lookups.md +24 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/js-tosorted-immutable.md +57 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-activity.md +26 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-animate-svg-wrapper.md +47 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-conditional-render.md +40 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-content-visibility.md +38 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-hoist-jsx.md +46 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-hydration-no-flicker.md +82 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-hydration-suppress-warning.md +30 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-svg-precision.md +28 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rendering-usetransition-loading.md +75 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-defer-reads.md +39 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-dependencies.md +45 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-derived-state-no-effect.md +40 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-derived-state.md +29 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-functional-setstate.md +74 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-lazy-state-init.md +58 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-memo-with-default-value.md +38 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-memo.md +44 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-move-effect-to-event.md +45 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-simple-expression-in-memo.md +35 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-transitions.md +40 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/rerender-use-ref-transient-values.md +73 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/server-after-nonblocking.md +73 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/server-auth-actions.md +96 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/server-cache-lru.md +41 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/server-cache-react.md +76 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/server-dedup-props.md +65 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/server-parallel-fetching.md +83 -0
- package/core/.agents/skills/vercel-react-best-practices/rules/server-serialization.md +38 -0
- package/core/.claude/agents/pcb-reviewer.md +146 -0
- package/core/.claude/commands/Amelia.md +5 -0
- package/core/.claude/commands/Ansuz.md +38 -0
- package/core/.claude/commands/Huginn.md +25 -0
- package/core/.claude/commands/agent-worker.md +168 -0
- package/core/.claude/commands/architect.md +24 -0
- package/core/.claude/commands/archive.md +24 -0
- package/core/.claude/commands/canary.md +17 -0
- package/core/.claude/commands/incident.md +15 -0
- package/core/.claude/commands/init.md +112 -0
- package/core/.claude/commands/legacy_dev.md +18 -0
- package/core/.claude/commands/product-review.md +131 -0
- package/core/.claude/commands/review.md +14 -0
- package/core/.claude/commands/security.md +15 -0
- package/core/.claude/commands/spec.md +22 -0
- package/core/.claude/commands/strategy.md +141 -0
- package/core/.claude/commands/sunny.md +1 -0
- package/core/.claude/commands//347/224/242/345/223/201.md +15 -0
- package/core/.claude/commands//347/224/242/345/223/201/346/252/242/350/246/226.md +129 -0
- package/core/.claude/commands//350/250/255/350/250/210.md +14 -0
- package/core/.claude/commands//350/251/225/344/274/260.md +16 -0
- package/core/.claude/commands//351/200/262/345/272/246.md +15 -0
- package/core/.claude/hooks/__tests__/substrate.test.js +122 -0
- package/core/.claude/hooks/__tests__/task-1866-cli-guard.test.js +193 -0
- package/core/.claude/hooks/__tests__/task-1977-dedupe-guard.test.js +125 -0
- package/core/.claude/hooks/__tests__/task-2087-mark-landed-idempotent.test.js +128 -0
- package/core/.claude/hooks/__tests__/task-2253-null-tip-marker-reuse.test.js +84 -0
- package/core/.claude/hooks/__tests__/task-2381-project-guard-walkup.test.js +53 -0
- package/core/.claude/hooks/_ansuz-conversation.js +183 -0
- package/core/.claude/hooks/_ansuz-conversation.test.js +90 -0
- package/core/.claude/hooks/_dedupe-guard.js +75 -0
- package/core/.claude/hooks/_peer-project-root.js +113 -0
- package/core/.claude/hooks/_peer-project-root.test.js +121 -0
- package/core/.claude/hooks/_peer-waiting-line.js +27 -0
- package/core/.claude/hooks/_project-guard.js +45 -0
- package/core/.claude/hooks/_run-user-prompt.js +134 -0
- package/core/.claude/hooks/_run-user-prompt.test.js +80 -0
- package/core/.claude/hooks/_session-identity.js +62 -0
- package/core/.claude/hooks/_session-pin.js +34 -0
- package/core/.claude/hooks/_session-pin.test.js +89 -0
- package/core/.claude/hooks/_signal-fence.js +17 -0
- package/core/.claude/hooks/_substrate-paths.js +95 -0
- package/core/.claude/hooks/_user-scope.js +23 -0
- package/core/.claude/hooks/ansuz-session-pin-via-turn-context.test.js +52 -0
- package/core/.claude/hooks/ansuz-session-pin.js +171 -0
- package/core/.claude/hooks/ansuz-session-pin.test.js +140 -0
- package/core/.claude/hooks/context-check.js +51 -0
- package/core/.claude/hooks/core-resolve.js +230 -0
- package/core/.claude/hooks/daemon-outbox-hook.js +139 -0
- package/core/.claude/hooks/fatigue-check.js +52 -0
- package/core/.claude/hooks/handoff-snapshot-on-commit.js +154 -0
- package/core/.claude/hooks/handoff-snapshot-on-decision.js +188 -0
- package/core/.claude/hooks/handoff-snapshot-on-question.js +158 -0
- package/core/.claude/hooks/handoff-staleness.js +197 -0
- package/core/.claude/hooks/jev-thalamus-context.js +17 -0
- package/core/.claude/hooks/jev-thalamus-context.test.js +41 -0
- package/core/.claude/hooks/lease-release.js +119 -0
- package/core/.claude/hooks/lease-touch.js +96 -0
- package/core/.claude/hooks/memory-guard.js +723 -0
- package/core/.claude/hooks/memory-guard.test.js +574 -0
- package/core/.claude/hooks/reflex-feedback.js +5 -0
- package/core/.claude/hooks/session-init.ps1 +49 -0
- package/core/.claude/hooks/sound-event.js +70 -0
- package/core/.claude/hooks/sound-event.sh +37 -0
- package/core/.claude/hooks/stream-delta.js +169 -0
- package/core/.claude/hooks/synapse-first-investigation.js +326 -0
- package/core/.claude/hooks/synapse-first-investigation.test.js +126 -0
- package/core/.claude/hooks/task-bridge.js +33 -0
- package/core/.claude/hooks/turn-context-peer.test.js +115 -0
- package/core/.claude/hooks/turn-context-session-line.test.js +69 -0
- package/core/.claude/hooks/turn-context.js +215 -0
- package/core/.claude/hooks/worktree-create.js +1232 -0
- package/core/.claude/hooks/worktree-remove.js +181 -0
- package/core/.claude/no-hooks.settings.json +3 -0
- package/core/.claude/settings.json +180 -0
- package/core/.claude/skills/adapt/SKILL.md +199 -0
- package/core/.claude/skills/animate/SKILL.md +190 -0
- package/core/.claude/skills/architect_agent/SKILL.md +408 -0
- package/core/.claude/skills/assessment_agent/SKILL.md +289 -0
- package/core/.claude/skills/audit/SKILL.md +129 -0
- package/core/.claude/skills/backend_agent/SKILL.md +407 -0
- package/core/.claude/skills/backend_engineer/SKILL.md +788 -0
- package/core/.claude/skills/bolder/SKILL.md +132 -0
- package/core/.claude/skills/clarify/SKILL.md +180 -0
- package/core/.claude/skills/code_reviewer/SKILL.md +305 -0
- package/core/.claude/skills/code_reviewer/patterns.yaml +85 -0
- package/core/.claude/skills/colorize/SKILL.md +158 -0
- package/core/.claude/skills/critique/SKILL.md +118 -0
- package/core/.claude/skills/delight/SKILL.md +317 -0
- package/core/.claude/skills/designer/SKILL.md +845 -0
- package/core/.claude/skills/dev_agent/SKILL.md +503 -0
- package/core/.claude/skills/developer/SKILL.md +792 -0
- package/core/.claude/skills/distill/SKILL.md +137 -0
- package/core/.claude/skills/extract/SKILL.md +95 -0
- package/core/.claude/skills/feature_agent/SKILL.md +303 -0
- package/core/.claude/skills/frontend-design/SKILL.md +127 -0
- package/core/.claude/skills/frontend-design/reference/color-and-contrast.md +132 -0
- package/core/.claude/skills/frontend-design/reference/interaction-design.md +123 -0
- package/core/.claude/skills/frontend-design/reference/motion-design.md +99 -0
- package/core/.claude/skills/frontend-design/reference/responsive-design.md +114 -0
- package/core/.claude/skills/frontend-design/reference/spatial-design.md +100 -0
- package/core/.claude/skills/frontend-design/reference/typography.md +131 -0
- package/core/.claude/skills/frontend-design/reference/ux-writing.md +107 -0
- package/core/.claude/skills/frontend_agent/SKILL.md +327 -0
- package/core/.claude/skills/harden/SKILL.md +358 -0
- package/core/.claude/skills/heimdall/SKILL.md +84 -0
- package/core/.claude/skills/huginn/SKILL.md +208 -0
- package/core/.claude/skills/incident_response/SKILL.md +315 -0
- package/core/.claude/skills/legacy_dev/SKILL.md +202 -0
- package/core/.claude/skills/normalize/SKILL.md +67 -0
- package/core/.claude/skills/onboard/SKILL.md +243 -0
- package/core/.claude/skills/optimization_agent/SKILL.md +340 -0
- package/core/.claude/skills/optimize/SKILL.md +269 -0
- package/core/.claude/skills/pcb-review/SKILL.md +39 -0
- package/core/.claude/skills/polish/SKILL.md +202 -0
- package/core/.claude/skills/product_manager/SKILL.md +1091 -0
- package/core/.claude/skills/qa_engineer/SKILL.md +1269 -0
- package/core/.claude/skills/quieter/SKILL.md +118 -0
- package/core/.claude/skills/remotion-best-practices/SKILL.md +61 -0
- package/core/.claude/skills/remotion-best-practices/rules/3d.md +86 -0
- package/core/.claude/skills/remotion-best-practices/rules/animations.md +27 -0
- package/core/.claude/skills/remotion-best-practices/rules/assets/charts-bar-chart.tsx +173 -0
- package/core/.claude/skills/remotion-best-practices/rules/assets/text-animations-typewriter.tsx +100 -0
- package/core/.claude/skills/remotion-best-practices/rules/assets/text-animations-word-highlight.tsx +103 -0
- package/core/.claude/skills/remotion-best-practices/rules/assets.md +78 -0
- package/core/.claude/skills/remotion-best-practices/rules/audio-visualization.md +198 -0
- package/core/.claude/skills/remotion-best-practices/rules/audio.md +169 -0
- package/core/.claude/skills/remotion-best-practices/rules/calculate-metadata.md +134 -0
- package/core/.claude/skills/remotion-best-practices/rules/can-decode.md +81 -0
- package/core/.claude/skills/remotion-best-practices/rules/charts.md +120 -0
- package/core/.claude/skills/remotion-best-practices/rules/compositions.md +154 -0
- package/core/.claude/skills/remotion-best-practices/rules/display-captions.md +184 -0
- package/core/.claude/skills/remotion-best-practices/rules/extract-frames.md +229 -0
- package/core/.claude/skills/remotion-best-practices/rules/ffmpeg.md +38 -0
- package/core/.claude/skills/remotion-best-practices/rules/fonts.md +152 -0
- package/core/.claude/skills/remotion-best-practices/rules/get-audio-duration.md +58 -0
- package/core/.claude/skills/remotion-best-practices/rules/get-video-dimensions.md +68 -0
- package/core/.claude/skills/remotion-best-practices/rules/get-video-duration.md +60 -0
- package/core/.claude/skills/remotion-best-practices/rules/gifs.md +141 -0
- package/core/.claude/skills/remotion-best-practices/rules/images.md +134 -0
- package/core/.claude/skills/remotion-best-practices/rules/import-srt-captions.md +69 -0
- package/core/.claude/skills/remotion-best-practices/rules/light-leaks.md +73 -0
- package/core/.claude/skills/remotion-best-practices/rules/lottie.md +70 -0
- package/core/.claude/skills/remotion-best-practices/rules/maps.md +412 -0
- package/core/.claude/skills/remotion-best-practices/rules/measuring-dom-nodes.md +34 -0
- package/core/.claude/skills/remotion-best-practices/rules/measuring-text.md +140 -0
- package/core/.claude/skills/remotion-best-practices/rules/parameters.md +109 -0
- package/core/.claude/skills/remotion-best-practices/rules/sequencing.md +118 -0
- package/core/.claude/skills/remotion-best-practices/rules/sfx.md +30 -0
- package/core/.claude/skills/remotion-best-practices/rules/subtitles.md +36 -0
- package/core/.claude/skills/remotion-best-practices/rules/tailwind.md +11 -0
- package/core/.claude/skills/remotion-best-practices/rules/text-animations.md +20 -0
- package/core/.claude/skills/remotion-best-practices/rules/timing.md +179 -0
- package/core/.claude/skills/remotion-best-practices/rules/transcribe-captions.md +70 -0
- package/core/.claude/skills/remotion-best-practices/rules/transitions.md +197 -0
- package/core/.claude/skills/remotion-best-practices/rules/transparent-videos.md +106 -0
- package/core/.claude/skills/remotion-best-practices/rules/trimming.md +51 -0
- package/core/.claude/skills/remotion-best-practices/rules/videos.md +171 -0
- package/core/.claude/skills/remotion-best-practices/rules/voiceover.md +99 -0
- package/core/.claude/skills/security_agent/SKILL.md +312 -0
- package/core/.claude/skills/spec_optimizer/SKILL.md +299 -0
- package/core/.claude/skills/sre_agent/SKILL.md +1347 -0
- package/core/.claude/skills/sunny/SKILL.md +74 -0
- package/core/.claude/skills/supabase-postgres-best-practices/AGENTS.md +90 -0
- package/core/.claude/skills/supabase-postgres-best-practices/CLAUDE.md +1 -0
- package/core/.claude/skills/supabase-postgres-best-practices/SKILL.md +64 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/advanced-full-text-search.md +55 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/advanced-jsonb-indexing.md +49 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/conn-idle-timeout.md +46 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/conn-limits.md +44 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/conn-pooling.md +41 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/conn-prepared-statements.md +46 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/data-batch-inserts.md +54 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/data-n-plus-one.md +53 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/data-pagination.md +50 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/data-upsert.md +50 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/lock-advisory.md +56 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/lock-deadlock-prevention.md +68 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/lock-short-transactions.md +50 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/lock-skip-locked.md +54 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/monitor-explain-analyze.md +45 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/monitor-pg-stat-statements.md +55 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/monitor-vacuum-analyze.md +55 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/query-composite-indexes.md +44 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/query-covering-indexes.md +40 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/query-index-types.md +45 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/query-missing-indexes.md +43 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/query-partial-indexes.md +45 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/schema-data-types.md +46 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/schema-foreign-key-indexes.md +59 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/schema-lowercase-identifiers.md +55 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/schema-partitioning.md +55 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/schema-primary-keys.md +61 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/security-privileges.md +54 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/security-rls-basics.md +50 -0
- package/core/.claude/skills/supabase-postgres-best-practices/references/security-rls-performance.md +57 -0
- package/core/.claude/skills/synapse_maintain/SKILL.md +84 -0
- package/core/.claude/skills/ui-ux-pro-max/SKILL.md +377 -0
- package/core/.claude/skills/ui-ux-pro-max/data/charts.csv +26 -0
- package/core/.claude/skills/ui-ux-pro-max/data/colors.csv +97 -0
- package/core/.claude/skills/ui-ux-pro-max/data/icons.csv +101 -0
- package/core/.claude/skills/ui-ux-pro-max/data/landing.csv +31 -0
- package/core/.claude/skills/ui-ux-pro-max/data/products.csv +97 -0
- package/core/.claude/skills/ui-ux-pro-max/data/react-performance.csv +45 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/astro.csv +54 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/flutter.csv +53 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/html-tailwind.csv +56 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/jetpack-compose.csv +53 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/nextjs.csv +53 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/nuxt-ui.csv +51 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/nuxtjs.csv +59 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/react-native.csv +52 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/react.csv +54 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/shadcn.csv +61 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/svelte.csv +54 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/swiftui.csv +51 -0
- package/core/.claude/skills/ui-ux-pro-max/data/stacks/vue.csv +50 -0
- package/core/.claude/skills/ui-ux-pro-max/data/styles.csv +68 -0
- package/core/.claude/skills/ui-ux-pro-max/data/typography.csv +58 -0
- package/core/.claude/skills/ui-ux-pro-max/data/ui-reasoning.csv +101 -0
- package/core/.claude/skills/ui-ux-pro-max/data/ux-guidelines.csv +100 -0
- package/core/.claude/skills/ui-ux-pro-max/data/web-interface.csv +31 -0
- package/core/.claude/skills/ui-ux-pro-max/scripts/core.py +253 -0
- package/core/.claude/skills/ui-ux-pro-max/scripts/design_system.py +1067 -0
- package/core/.claude/skills/ui-ux-pro-max/scripts/search.py +106 -0
- package/core/.claude/skills/vercel-react-best-practices/AGENTS.md +2934 -0
- package/core/.claude/skills/vercel-react-best-practices/SKILL.md +136 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/advanced-event-handler-refs.md +55 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/advanced-init-once.md +42 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/advanced-use-latest.md +39 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/async-api-routes.md +38 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/async-defer-await.md +80 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/async-dependencies.md +51 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/async-parallel.md +28 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/async-suspense-boundaries.md +99 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/bundle-barrel-imports.md +59 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/bundle-conditional.md +31 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/bundle-defer-third-party.md +49 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/bundle-dynamic-imports.md +35 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/bundle-preload.md +50 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/client-event-listeners.md +74 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/client-localstorage-schema.md +71 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/client-passive-event-listeners.md +48 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/client-swr-dedup.md +56 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-batch-dom-css.md +107 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-cache-function-results.md +80 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-cache-property-access.md +28 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-cache-storage.md +70 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-combine-iterations.md +32 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-early-exit.md +50 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-hoist-regexp.md +45 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-index-maps.md +37 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-length-check-first.md +49 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-min-max-loop.md +82 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-set-map-lookups.md +24 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/js-tosorted-immutable.md +57 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-activity.md +26 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-animate-svg-wrapper.md +47 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-conditional-render.md +40 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-content-visibility.md +38 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-hoist-jsx.md +46 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-hydration-no-flicker.md +82 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-hydration-suppress-warning.md +30 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-svg-precision.md +28 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rendering-usetransition-loading.md +75 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-defer-reads.md +39 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-dependencies.md +45 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-derived-state-no-effect.md +40 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-derived-state.md +29 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-functional-setstate.md +74 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-lazy-state-init.md +58 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-memo-with-default-value.md +38 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-memo.md +44 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-move-effect-to-event.md +45 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-simple-expression-in-memo.md +35 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-transitions.md +40 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/rerender-use-ref-transient-values.md +73 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/server-after-nonblocking.md +73 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/server-auth-actions.md +96 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/server-cache-lru.md +41 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/server-cache-react.md +76 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/server-dedup-props.md +65 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/server-parallel-fetching.md +83 -0
- package/core/.claude/skills/vercel-react-best-practices/rules/server-serialization.md +38 -0
- package/core/.collaboration/README.md +258 -0
- package/core/.collaboration/cli/commands/assign.js +39 -0
- package/core/.collaboration/cli/commands/claim.js +33 -0
- package/core/.collaboration/cli/commands/complete.js +44 -0
- package/core/.collaboration/cli/commands/create.js +110 -0
- package/core/.collaboration/cli/commands/get.js +83 -0
- package/core/.collaboration/cli/commands/index.js +52 -0
- package/core/.collaboration/cli/commands/list.js +51 -0
- package/core/.collaboration/cli/commands/lock.js +45 -0
- package/core/.collaboration/cli/commands/rebuild.js +39 -0
- package/core/.collaboration/cli/commands/search.js +61 -0
- package/core/.collaboration/cli/commands/stats.js +61 -0
- package/core/.collaboration/cli/commands/transfer.js +41 -0
- package/core/.collaboration/cli/commands/unlock.js +48 -0
- package/core/.collaboration/cli/formatters/help-formatter.js +70 -0
- package/core/.collaboration/cli/formatters/index.js +7 -0
- package/core/.collaboration/cli/task-cli.js +46 -0
- package/core/.collaboration/cli/utils/arg-parser.js +74 -0
- package/core/.collaboration/cli/utils/git-user.js +25 -0
- package/core/.collaboration/cli/utils/index.js +9 -0
- package/core/.collaboration/cli/utils/output.js +10 -0
- package/core/.collaboration/config/default.yaml +189 -0
- package/core/.collaboration/core/conflict-resolver.js +461 -0
- package/core/.collaboration/core/conflict-resolver.test.js +389 -0
- package/core/.collaboration/core/id-generator.js +303 -0
- package/core/.collaboration/core/migrate-legacy.js +260 -0
- package/core/.collaboration/core/package.json +1 -0
- package/core/.collaboration/core/task-2257-estimated-hours.test.js +72 -0
- package/core/.collaboration/core/task-2260-dir-layout-move.test.js +60 -0
- package/core/.collaboration/core/task-2264-single-after-capture.test.js +61 -0
- package/core/.collaboration/core/task-2288-followups.test.js +121 -0
- package/core/.collaboration/core/task-api-json.js +211 -0
- package/core/.collaboration/core/task-api.js +3729 -0
- package/core/.collaboration/core/task-api.test.js +1268 -0
- package/core/.collaboration/docs/API.md +787 -0
- package/core/.collaboration/docs/ARCHITECTURE.md +596 -0
- package/core/.collaboration/docs/CLI.md +821 -0
- package/core/.collaboration/docs/FAQ.md +695 -0
- package/core/.collaboration/docs/MIGRATION.md +804 -0
- package/core/.collaboration/docs/examples/README.md +320 -0
- package/core/.collaboration/docs/examples/advanced-queries.js +261 -0
- package/core/.collaboration/docs/examples/basic-usage.js +184 -0
- package/core/.collaboration/docs/examples/team-workflow.js +352 -0
- package/core/.collaboration/hooks/README.md +70 -0
- package/core/.collaboration/hooks/config.json +10 -0
- package/core/.collaboration/hooks/install.js +400 -0
- package/core/.collaboration/hooks/post-merge.js +163 -0
- package/core/.collaboration/hooks/pre-commit.js +264 -0
- package/core/.collaboration/hooks/synapse-hook.js +486 -0
- package/core/.collaboration/hooks/synapse-hook.test.js +69 -0
- package/core/.failure/INDEX.yaml +231 -0
- package/core/.failure/agent-execution/FAIL-001-agent-no-output.yaml +391 -0
- package/core/.failure/agent-execution/FAIL-004_agent_platform_specific_commands.md +194 -0
- package/core/.failure/agent-execution/FAIL-005_background_agent_permission_denied.md +204 -0
- package/core/.failure/agent-execution/FAIL-007-pm-lead-wrong-team-invocation.yaml +0 -0
- package/core/.failure/agent-execution/FAIL-008-bypass-api-direct-yaml-write.yaml +94 -0
- package/core/.failure/agent-execution/FAIL-010-phase0-poem-format-skipped.yaml +28 -0
- package/core/.failure/agent-execution/FAIL-010-skipping-cognitive-checkpoint-after-verification.yaml +59 -0
- package/core/.failure/agent-execution/FAIL-012-grep-shortcut-during-investigation.yaml +85 -0
- package/core/.failure/agent-execution/FAIL-013-search-cwd-drift-and-case-mismatch.yaml +103 -0
- package/core/.failure/agent-execution/FAIL-014-shared-worktree-premature-death-verdict.yaml +35 -0
- package/core/.failure/agent-execution/FAIL-AUTO-TASK_1413.yaml +31 -0
- package/core/.failure/agent-execution/FAIL-AUTO-TASK_1427.yaml +41 -0
- package/core/.failure/agent-execution/FAIL-AUTO-TASK_904.yaml +30 -0
- package/core/.failure/agent-execution/README.md +68 -0
- package/core/.failure/architecture/FAIL-004_session_based_skill_recognition.md +128 -0
- package/core/.failure/architecture/FAIL-006-python-version-compatibility.yaml +445 -0
- package/core/.failure/architecture/README.md +87 -0
- package/core/.failure/code-quality/README.md +89 -0
- package/core/.failure/data-integrity/FAIL-002-json-bom.yaml +220 -0
- package/core/.failure/data-integrity/README.md +141 -0
- package/core/.failure/pipeline/FAIL-003-queue-pollution.yaml +335 -0
- package/core/.failure/pipeline/README.md +95 -0
- package/core/.failure/retired/FAIL-009-skipping-synapse-pulse-before-code-change.yaml +49 -0
- package/core/.failure/tools/README.md +230 -0
- package/core/.failure/tools/USAGE_GUIDE.md +323 -0
- package/core/.failure/tools/check_bom.js +101 -0
- package/core/.failure/tools/check_failures.ps1 +387 -0
- package/core/.failure/tools/fix_json.js +58 -0
- package/core/.failure/tools/log_failure.ps1 +474 -0
- package/core/.failure/tools/remove_bom.js +46 -0
- package/core/.failure/tools/validate_json.js +50 -0
- package/core/.formation/.episode-input.json +10 -0
- package/core/.formation/README.md +188 -0
- package/core/.formation/bifrost/CGDD_hook_lifecycle.js +575 -0
- package/core/.formation/bifrost/HOOK_PARITY_VERIFICATION.md +241 -0
- package/core/.formation/bifrost/README.md +71 -0
- package/core/.formation/bifrost/codex-doctor.js +106 -0
- package/core/.formation/bifrost/codex-doctor.test.js +51 -0
- package/core/.formation/bifrost/launch.js +510 -0
- package/core/.formation/bifrost/launch.test.js +217 -0
- package/core/.formation/bifrost/openai/codex-hook-shim.js +171 -0
- package/core/.formation/bifrost/openai/codex-hook-shim.test.js +107 -0
- package/core/.formation/bifrost/openai/models.json +36 -0
- package/core/.formation/bifrost/openai/relay.js +781 -0
- package/core/.formation/bifrost/openai/relay.test.js +71 -0
- package/core/.formation/bifrost/session.js +382 -0
- package/core/.formation/bifrost/session.test.js +121 -0
- package/core/.formation/bifrost/setup-codex.js +685 -0
- package/core/.formation/bifrost/setup-codex.test.js +102 -0
- package/core/.formation/bifrost/shared/ansuz-boot-policy.js +31 -0
- package/core/.formation/bifrost/shared/ansuz-boot-policy.test.js +40 -0
- package/core/.formation/bifrost/shared/boot-context.js +68 -0
- package/core/.formation/bifrost/shared/boot-context.test.js +67 -0
- package/core/.formation/bifrost/shared/codex-appserver.js +191 -0
- package/core/.formation/bifrost/shared/codex-injection.js +126 -0
- package/core/.formation/bifrost/shared/codex-session-registry.js +105 -0
- package/core/.formation/bifrost/shared/codex-session-registry.test.js +126 -0
- package/core/.formation/bifrost/shared/consolidate.js +286 -0
- package/core/.formation/bifrost/shared/stream-json.js +103 -0
- package/core/.formation/bifrost/shared/tools.js +278 -0
- package/core/.formation/bifrost/sync.js +142 -0
- package/core/.formation/briefs/scheduler-evolution.md +209 -0
- package/core/.formation/lead.yaml +118 -0
- package/core/.formation/legacy.bak/designer.yaml +57 -0
- package/core/.formation/legacy.bak/pm.yaml +57 -0
- package/core/.formation/legacy.bak/qa.yaml +57 -0
- package/core/.formation/legacy.bak/review.yaml +62 -0
- package/core/.formation/legacy.bak/scheduler.yaml +129 -0
- package/core/.formation/legacy.bak/sre.yaml +57 -0
- package/core/.formation/legion/forge.yaml +299 -0
- package/core/.formation/legion/hive.yaml +238 -0
- package/core/.formation/native/claude/README.md +61 -0
- package/core/.formation/native/claude/models.json +42 -0
- package/core/.formation/role-map.yaml +274 -0
- package/core/.formation/roles/_shared/modules/code_review.yaml +98 -0
- package/core/.formation/roles/_shared/modules/security.yaml +103 -0
- package/core/.formation/roles/_shared/modules/testing.yaml +14 -0
- package/core/.formation/roles/backend/role.yaml +93 -0
- package/core/.formation/roles/code_reviewer/role.yaml +102 -0
- package/core/.formation/roles/database_optimizer/role.yaml +107 -0
- package/core/.formation/roles/designer/role.yaml +66 -0
- package/core/.formation/roles/developer/role.yaml +75 -0
- package/core/.formation/roles/frontend/role.yaml +134 -0
- package/core/.formation/roles/incident_response/role.yaml +89 -0
- package/core/.formation/roles/qa/role.yaml +89 -0
- package/core/.formation/roles/security/role.yaml +87 -0
- package/core/.formation/roles/sre/role.yaml +66 -0
- package/core/.formation/skills/coordinator.yaml +52 -0
- package/core/.formation/solo/diagnosis.yaml +78 -0
- package/core/.formation/solo/strategy.yaml +121 -0
- package/core/.formation/spec/task_operations.parity.test.js +426 -0
- package/core/.formation/spec/task_operations.yaml +291 -0
- package/core/.formation/squad/blitz.yaml +192 -0
- package/core/.formation/squad/strategy-tribunal.yaml +354 -0
- package/core/.formation/squad/strategy.yaml +349 -0
- package/core/.formation/squad/sweep.yaml +208 -0
- package/core/.formation/tests/hello.txt +2 -0
- package/core/.formation/tools/__tests__/m1-claim-gate.test.js +175 -0
- package/core/.formation/tools/__tests__/m1-hmap-convid.test.js +216 -0
- package/core/.formation/tools/__tests__/release-lock.test.js +176 -0
- package/core/.formation/tools/__tests__/subagent-bifrost-guard.test.js +234 -0
- package/core/.formation/tools/__tests__/subagent-codex-harness.test.js +96 -0
- package/core/.formation/tools/__tests__/task-1863-claim-late-release.test.js +160 -0
- package/core/.formation/tools/__tests__/task-2265-singleton-identity.test.js +134 -0
- package/core/.formation/tools/__tests__/task-2281-singleton-readers.test.js +64 -0
- package/core/.formation/tools/__tests__/task2030-worktree-node-modules.test.js +221 -0
- package/core/.formation/tools/__tests__/task2067-worktree-branch-reuse.test.js +148 -0
- package/core/.formation/tools/alaya-gate.js +198 -0
- package/core/.formation/tools/analytics.js +172 -0
- package/core/.formation/tools/broadcast.js +103 -0
- package/core/.formation/tools/commit-sync.js +154 -0
- package/core/.formation/tools/coordinator.js +350 -0
- package/core/.formation/tools/dashboard-models-chain.test.js +60 -0
- package/core/.formation/tools/dispatch-cortex-context.test.js +52 -0
- package/core/.formation/tools/dispatch-detail/cross-project-worktree.md +40 -0
- package/core/.formation/tools/dispatch-detail/testing-module.md +53 -0
- package/core/.formation/tools/dispatch-detail/worktree-subagent-mode.md +72 -0
- package/core/.formation/tools/dispatch-dryrun.test.js +54 -0
- package/core/.formation/tools/dispatch-model-policy.test.js +165 -0
- package/core/.formation/tools/dispatch-notifier.js +84 -0
- package/core/.formation/tools/dispatch-registry.js +346 -0
- package/core/.formation/tools/dispatch.js +1931 -0
- package/core/.formation/tools/formations.js +1203 -0
- package/core/.formation/tools/forseti-verify.js +1170 -0
- package/core/.formation/tools/git-salvage.js +269 -0
- package/core/.formation/tools/guard-attribution.js +224 -0
- package/core/.formation/tools/headless-branch-recovery.test.js +64 -0
- package/core/.formation/tools/headless-dryrun.test.js +79 -0
- package/core/.formation/tools/headless-v2.js +659 -0
- package/core/.formation/tools/headless.js +1071 -0
- package/core/.formation/tools/integration.js +528 -0
- package/core/.formation/tools/jev-correction.js +214 -0
- package/core/.formation/tools/jev-correction.test.js +178 -0
- package/core/.formation/tools/legacy-scheduler.js +1994 -0
- package/core/.formation/tools/lib/exec-tree.js +92 -0
- package/core/.formation/tools/lib/exec-tree.test.js +82 -0
- package/core/.formation/tools/loader.js +190 -0
- package/core/.formation/tools/log-reader.js +232 -0
- package/core/.formation/tools/logger.js +273 -0
- package/core/.formation/tools/mcp-allowlist.test.js +40 -0
- package/core/.formation/tools/memory-loader.js +422 -0
- package/core/.formation/tools/merge-lock-heartbeat.test.js +35 -0
- package/core/.formation/tools/no-hooks.settings.json +3 -0
- package/core/.formation/tools/pool-status-push.js +121 -0
- package/core/.formation/tools/pool-status-push.test.js +165 -0
- package/core/.formation/tools/post-complete/bounded-http.js +102 -0
- package/core/.formation/tools/post-complete/bounded-http.test.js +130 -0
- package/core/.formation/tools/post-complete/cognitive.js +270 -0
- package/core/.formation/tools/post-complete/git-merge.js +2845 -0
- package/core/.formation/tools/post-complete/health.js +464 -0
- package/core/.formation/tools/post-complete/integration-gaps.js +98 -0
- package/core/.formation/tools/post-complete/outcome.js +874 -0
- package/core/.formation/tools/post-complete/staging.js +390 -0
- package/core/.formation/tools/post-complete/subagent-transcript-stats.js +460 -0
- package/core/.formation/tools/post-complete/subagent-transcript-stats.test.js +680 -0
- package/core/.formation/tools/post-complete/task1536.test.js +627 -0
- package/core/.formation/tools/post-complete/task1543.test.js +261 -0
- package/core/.formation/tools/post-complete/task1876.test.js +287 -0
- package/core/.formation/tools/post-complete/task1957.test.js +293 -0
- package/core/.formation/tools/post-complete/task1997.test.js +401 -0
- package/core/.formation/tools/post-complete/task2028.test.js +232 -0
- package/core/.formation/tools/post-complete/task2029.test.js +275 -0
- package/core/.formation/tools/post-complete/task2036.test.js +330 -0
- package/core/.formation/tools/post-complete/task2038.test.js +246 -0
- package/core/.formation/tools/post-complete/task2071.test.js +231 -0
- package/core/.formation/tools/post-complete/task2233.test.js +100 -0
- package/core/.formation/tools/post-complete/task2240.test.js +134 -0
- package/core/.formation/tools/post-complete/task2244.test.js +199 -0
- package/core/.formation/tools/post-complete/task2255.test.js +28 -0
- package/core/.formation/tools/post-complete/task2268.test.js +106 -0
- package/core/.formation/tools/post-complete/task2341.test.js +74 -0
- package/core/.formation/tools/post-complete/task2345.test.js +38 -0
- package/core/.formation/tools/post-complete/testing.js +482 -0
- package/core/.formation/tools/post-complete/wish.js +177 -0
- package/core/.formation/tools/post-complete/worktree-rescue.test.js +283 -0
- package/core/.formation/tools/post-complete.js +3660 -0
- package/core/.formation/tools/post-complete.test.js +1002 -0
- package/core/.formation/tools/pre-dispatch-snapshot.test.js +328 -0
- package/core/.formation/tools/process-state.js +33 -0
- package/core/.formation/tools/prompt.js +2183 -0
- package/core/.formation/tools/prompt.test.js +875 -0
- package/core/.formation/tools/remote-conflict.js +98 -0
- package/core/.formation/tools/revert-daemon-state.js +162 -0
- package/core/.formation/tools/route-audit.test.js +55 -0
- package/core/.formation/tools/runner.js +1332 -0
- package/core/.formation/tools/supabase-client.js +104 -0
- package/core/.formation/tools/task-sync.js +236 -0
- package/core/.formation/tools/task-utils.js +344 -0
- package/core/.formation/tools/task2028-headless-worktree.test.js +112 -0
- package/core/.formation/tools/test-drift.js +148 -0
- package/core/.formation/tools/tests/headless-mcp-args-plumbing.test.js +123 -0
- package/core/.formation/tools/tests/relay-mcp-overrides.test.js +112 -0
- package/core/.formation/tools/transcript-recorder.js +574 -0
- package/core/.formation/tools/utils-claude-cmd.test.js +28 -0
- package/core/.formation/tools/utils-worktree-config.test.js +44 -0
- package/core/.formation/tools/utils.js +1020 -0
- package/core/.formation/tools/verify-acceptance.js +585 -0
- package/core/.formation/tools/worktree-leak.test.js +162 -0
- package/core/.formation/tools/worktree-reclaim.js +259 -0
- package/core/.formation/tools/worktree-reclaim.test.js +182 -0
- package/core/.formation/tools/yggdrasil-local.js +97 -0
- package/core/.formation/unit/do.yaml +78 -0
- package/core/.formation/unit/heimdall.yaml +96 -0
- package/core/.formation/unit/plan.yaml +67 -0
- package/core/.formation/unit/think.yaml +58 -0
- package/core/.kenaz/MUNINN_SNAPSHOT9_EPISODE.json +32 -0
- package/core/.kenaz/MUNINN_SNAPSHOT9_RUNE.json +23 -0
- package/core/.kenaz/MUNINN_SNAPSHOT9_SUMMARY.md +89 -0
- package/core/.kenaz/ash_TASK_1428-20260507.json +14 -0
- package/core/.kenaz/cache/native-models/claude.json +42 -0
- package/core/.kenaz/claude-hooks-manifest.json +119 -0
- package/core/.kenaz/cortex.yaml +100 -0
- package/core/.kenaz/curator_state.json +7 -0
- package/core/.kenaz/dashboard-quick-prompts.json +9 -0
- package/core/.kenaz/designs/METAPLASTICITY_CONFIDENCE_DESIGN.md +537 -0
- package/core/.kenaz/designs/REFLEX_INITIATOR_DESIGN.md +501 -0
- package/core/.kenaz/designs/TASK_1385_handoff_mcp_design.md +426 -0
- package/core/.kenaz/designs/TASK_932_hippocampus_consolidation.md +103 -0
- package/core/.kenaz/designs/TASK_934_efe_signal_engine.md +97 -0
- package/core/.kenaz/designs/TASK_935_self_evolution_framework.md +154 -0
- package/core/.kenaz/designs/VISION_cognitive_architecture_roadmap.md +159 -0
- package/core/.kenaz/designs/metaplasticity_simulation_results.json +696 -0
- package/core/.kenaz/docs/INDEX.md +214 -0
- package/core/.kenaz/docs/README.md +290 -0
- package/core/.kenaz/docs/registry.json +962 -0
- package/core/.kenaz/docs/schema.json +144 -0
- package/core/.kenaz/docs/tools/GIT_HOOK_GUIDE.md +203 -0
- package/core/.kenaz/docs/tools/RAG_INTEGRATION_GUIDE.md +306 -0
- package/core/.kenaz/docs/tools/README.md +99 -0
- package/core/.kenaz/docs/tools/TIME_ESTIMATION_GUIDE.md +299 -0
- package/core/.kenaz/docs/tools/package-lock.json +618 -0
- package/core/.kenaz/docs/tools/package.json +34 -0
- package/core/.kenaz/docs/tools/src/cli.ts +365 -0
- package/core/.kenaz/docs/tools/src/field-extractor.ts +152 -0
- package/core/.kenaz/docs/tools/src/index.ts +14 -0
- package/core/.kenaz/docs/tools/src/registry.ts +664 -0
- package/core/.kenaz/docs/tools/src/search.ts +195 -0
- package/core/.kenaz/docs/tools/src/types.ts +88 -0
- package/core/.kenaz/docs/tools/test-git-hook.sh +79 -0
- package/core/.kenaz/docs/tools/tsconfig.json +20 -0
- package/core/.kenaz/docs/tools/verify.sh +106 -0
- package/core/.kenaz/episodes/SESSION_ansuz-prompt-1785087304784.yaml +68 -0
- package/core/.kenaz/huginn-exploration.json +8 -0
- package/core/.kenaz/knowledge/ladybug_2c51da1e_shadow_file_frame_group.patch +163 -0
- package/core/.kenaz/knowledge/ladybug_30cf85690e_shadow_page_read_during_checkpoint.patch +393 -0
- package/core/.kenaz/knowledge/ladybug_7a17858d3a_node_group_delete_lock.patch +102 -0
- package/core/.kenaz/knowledge/ladybug_926a0ba1_column_checkpoint.patch +412 -0
- package/core/.kenaz/knowledge/ladybug_a346c4c4b6_pk_index_header_pages.patch +144 -0
- package/core/.kenaz/knowledge/ladybug_ba5f38815b_csr_node_group.patch +86 -0
- package/core/.kenaz/main-memory/README.md +230 -0
- package/core/.kenaz/main-memory/kenaz/cortex/001_cortex_overview.md +168 -0
- package/core/.kenaz/main-memory/kenaz/cortex/001_rag_engine_overview.md +30 -0
- package/core/.kenaz/main-memory/kenaz/memory-system/001_three_tier_architecture.md +361 -0
- package/core/.kenaz/main-memory/kenaz/pipeline/001_task_execution_flow.md +117 -0
- package/core/.kenaz/main-memory/kenaz/pipeline/001_task_granularity.md +154 -0
- package/core/.kenaz/main-memory/kenaz/pipeline/002_team_priority.md +51 -0
- package/core/.kenaz/main-memory/kenaz/pipeline/003_feature_id_system.md +46 -0
- package/core/.kenaz/main-memory/kenaz/rules/001_gen_007_self_containment.md +108 -0
- package/core/.kenaz/main-memory/kenaz/rules/001_rule_priority_system.md +121 -0
- package/core/.kenaz/main-memory/kenaz/rules/002_agt_006_pre_check.md +0 -0
- package/core/.kenaz/main-memory/kenaz/rules/003_ppl_002_task_archival.md +48 -0
- package/core/.kenaz/main-memory/kenaz/system/001_dual_track_task_system.md +197 -0
- package/core/.kenaz/main-memory/knowledge/patterns/001_json_bom_issue.md +78 -0
- package/core/.kenaz/main-memory/knowledge/patterns/001_leader_worker_pattern.md +0 -0
- package/core/.kenaz/main-memory/knowledge/patterns/002_scope_path_conflict.md +115 -0
- package/core/.kenaz/main-memory/knowledge/patterns/002_task_dependency_resolution.md +223 -0
- package/core/.kenaz/main-memory/knowledge/patterns/003_question_protocol.md +242 -0
- package/core/.kenaz/main-memory/knowledge/patterns/004_user_insight_value.md +111 -0
- package/core/.kenaz/main-memory/knowledge/patterns/005_file_lock_pattern.md +274 -0
- package/core/.kenaz/main-memory/knowledge/references/001_ppl_rules.md +28 -0
- package/core/.kenaz/main-memory/knowledge/solutions/001_json_bom_issue.md +137 -0
- package/core/.kenaz/main-memory/knowledge/solutions/001_team_invocation.md +0 -0
- package/core/.kenaz/main-memory/knowledge/solutions/002_passive_monitoring.md +0 -0
- package/core/.kenaz/main-memory/knowledge/solutions/003_windows_file_race_condition.md +239 -0
- package/core/.kenaz/main-memory/registry.json +326 -0
- package/core/.kenaz/main-memory/version.yaml +58 -0
- package/core/.kenaz/muninn_episode_write.ps1 +17 -0
- package/core/.kenaz/muninn_snapshot11_rune_candidates.md +63 -0
- package/core/.kenaz/muninn_snapshot1_rune_candidates.md +49 -0
- package/core/.kenaz/orphan_audit_2026-05.md +181 -0
- package/core/.kenaz/ownership.yaml +155 -0
- package/core/.kenaz/peer.yaml +72 -0
- package/core/.kenaz/project.yaml +62 -0
- package/core/.kenaz/rune_ansuz_literature_validation_strategy.md +40 -0
- package/core/.kenaz/rune_efe_active_inference_validation.md +36 -0
- package/core/.kenaz/rune_yggdrasil_filesystem_scoped_project.md +41 -0
- package/core/.kenaz/rune_yggdrasil_project_identity_folder_name.md +41 -0
- package/core/.kenaz/rune_yggdrasil_relay_protocol_project_mismatch.md +39 -0
- package/core/.kenaz/runes_TASK_1428.json +57 -0
- package/core/.kenaz/specialist/ansuz_motivational_state.schema.json +138 -0
- package/core/.kenaz/specialist/ansuz_rules.yaml +241 -0
- package/core/.kenaz/specialist/ansuz_session_handoff.schema.json +128 -0
- package/core/.kenaz/specialist/chart_TASK_1657.json +50 -0
- package/core/.kenaz/specialist/cognitive-hub-e2e.test.js +494 -0
- package/core/.kenaz/specialist/daemon-manifest.yaml +168 -0
- package/core/.kenaz/specialist/empty_ac_audit_report.md +63 -0
- package/core/.kenaz/specialist/episode_muninn_completion_1785056865691.json +91 -0
- package/core/.kenaz/specialist/field_report_watermark.json +1 -0
- package/core/.kenaz/specialist/gc1826_driver.js +118 -0
- package/core/.kenaz/specialist/heimdall_proposal_1017c07f.json +13 -0
- package/core/.kenaz/specialist/heimdall_proposal_3dfaa8b9.json +12 -0
- package/core/.kenaz/specialist/heimdall_proposal_57af18f9.json +15 -0
- package/core/.kenaz/specialist/heimdall_proposal_57b1aa1b.json +14 -0
- package/core/.kenaz/specialist/heimdall_proposal_5a3c9473.json +13 -0
- package/core/.kenaz/specialist/identity_verdicts/TASK_1073.json +75 -0
- package/core/.kenaz/specialist/identity_verdicts/TASK_TEST.json +77 -0
- package/core/.kenaz/specialist/memory_guard_fired.json +1 -0
- package/core/.kenaz/specialist/memory_guard_trigger_cache.json +1 -0
- package/core/.kenaz/specialist/memory_source_priority.yaml +260 -0
- package/core/.kenaz/specialist/motivational_state.schema.json +62 -0
- package/core/.kenaz/specialist/prompt_1775.txt +317 -0
- package/core/.kenaz/specialist/prompt_1776.txt +326 -0
- package/core/.kenaz/specialist/prompt_1791.txt +320 -0
- package/core/.kenaz/specialist/prompt_1792.txt +320 -0
- package/core/.kenaz/specialist/prompt_1795.txt +281 -0
- package/core/.kenaz/specialist/prompt_1813.txt +332 -0
- package/core/.kenaz/specialist/prompt_1814.txt +401 -0
- package/core/.kenaz/specialist/prompt_1815.txt +394 -0
- package/core/.kenaz/specialist/prompt_1816.txt +437 -0
- package/core/.kenaz/specialist/prompt_1818.txt +361 -0
- package/core/.kenaz/specialist/prompt_1819.txt +421 -0
- package/core/.kenaz/specialist/prompt_1821.txt +345 -0
- package/core/.kenaz/specialist/prompt_1822.txt +313 -0
- package/core/.kenaz/specialist/prompt_1824.txt +348 -0
- package/core/.kenaz/specialist/prompt_1826.txt +388 -0
- package/core/.kenaz/specialist/retired_organs/README.md +58 -0
- package/core/.kenaz/specialist/rune_calibration_poc.md +175 -0
- package/core/.kenaz/specialist/rune_efe_salience_gap.md +21 -0
- package/core/.kenaz/specialist/rune_ember_autonomous_dispatch_gate.md +24 -0
- package/core/.kenaz/specialist/rune_post_dispatch_verification_pattern.md +31 -0
- package/core/.kenaz/specialist/rune_surprise_proxy_improvement.md +28 -0
- package/core/.kenaz/specialist/rune_worktree_dispatch_pattern.md +31 -0
- package/core/.kenaz/specialist/rune_yaml_filter_desync_fix.md +26 -0
- package/core/.kenaz/specialist/rune_yggdrasil_project_identity_folder_name.md +24 -0
- package/core/.kenaz/specialist/sessions/chart_TASK_1599.json +41 -0
- package/core/.kenaz/specialist/sessions/hmap_028aa781-aad2-48c0-9a61-5540565d4a7f.json +5 -0
- package/core/.kenaz/specialist/sessions/hmap_20210387-0c64-400c-9d0a-53042a0cbd89.json +5 -0
- package/core/.kenaz/specialist/sessions/hmap_c1375c5d-fd9f-44bd-a83e-353ecbb0392f.json +5 -0
- package/core/.kenaz/specialist/sessions/scratch_028aa781-aad2-48c0-9a61-5540565d4a7f.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_0acc0fb3-24f9-453f-9e92-6ef1cdf03a7c.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_20210387-0c64-400c-9d0a-53042a0cbd89.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_2f10623e-b7a1-4647-905e-9dc7c4d99f9c.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_5a38b62d-608a-488c-82f0-7e47e95e0905.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_5d9758ef-d9b6-4ffe-a017-3ed26d0a313b.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_c1375c5d-fd9f-44bd-a83e-353ecbb0392f.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_c9f3c711-b73d-4fcd-ad2a-7b2bc024ee77.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_e72e5274-1c81-4655-a05a-f87d1ecb2917.json +6 -0
- package/core/.kenaz/specialist/sessions/scratch_f740cdbe-f756-41fc-98fb-2b1413d51901.json +6 -0
- package/core/.kenaz/specialist/t1777_probe.txt +1 -0
- package/core/.kenaz/specialist/task1825_update.json +43 -0
- package/core/.kenaz/specialist/task1827_update.json +43 -0
- package/core/.kenaz/specialist/task1828_update.json +43 -0
- package/core/.kenaz/specialist/taskgen/consolidation_i.json +38 -0
- package/core/.kenaz/specialist/taskgen/csharp_a.json +44 -0
- package/core/.kenaz/specialist/taskgen/csharp_b.json +40 -0
- package/core/.kenaz/specialist/taskgen/embedding_e.json +34 -0
- package/core/.kenaz/specialist/taskgen/hygiene_fg.json +39 -0
- package/core/.kenaz/specialist/taskgen/pathology_h.json +36 -0
- package/core/.kenaz/specialist/taskgen/unity_noise_d.json +38 -0
- package/core/.kenaz/specialist/taskgen/visibility_c.json +38 -0
- package/core/.kenaz/specialist/verify_1837.js +302 -0
- package/core/.kenaz/specialist/verify_1838.js +124 -0
- package/core/.kenaz/tmp/heimdall_proposal_fc7e0944.json +10 -0
- package/core/.kenaz/tmp_heimdall_proposal_8d3e392.json +13 -0
- package/core/.kenaz/tmp_props.json +1 -0
- package/core/.kenaz/tmp_sounddrop_t1_prompt.txt +371 -0
- package/core/.kenaz/user_model.schema.json +107 -0
- package/core/.matrix/README.md +257 -0
- package/core/.matrix/index.matrix +5 -0
- package/core/.mcp.json +8 -0
- package/core/.prefrontal/.gitkeep +0 -0
- package/core/.prefrontal/proposals/.gitkeep +0 -0
- package/core/.rule/INDEX.yaml +99 -0
- package/core/.rule/MECHANIZED.md +61 -0
- package/core/.rule/code/README.md +211 -0
- package/core/.rule/code/backend.yaml +294 -0
- package/core/.rule/code/checklist.yaml +258 -0
- package/core/.rule/code/maintainability.yaml +227 -0
- package/core/.rule/code/naming.yaml +106 -0
- package/core/.rule/code/principles.yaml +71 -0
- package/core/.rule/code/reusability.yaml +192 -0
- package/core/.rule/code/structure.yaml +218 -0
- package/core/.rule/code/style.yaml +197 -0
- package/core/.rule/design/ui.yaml +79 -0
- package/core/.rule/general/agent-workflow.yaml +572 -0
- package/core/.rule/general/cortex-memory-recall.yaml +446 -0
- package/core/.rule/general/cortex-tools.yaml +181 -0
- package/core/.rule/general/cross-platform.yaml +107 -0
- package/core/.rule/general/ddx-synapse-stack.yaml +162 -0
- package/core/.rule/general/file-size-threshold.yaml +45 -0
- package/core/.rule/general/identity-contract.schema.yaml +251 -0
- package/core/.rule/general/memory-language.yaml +67 -0
- package/core/.rule/general/memory-source-priority.yaml +207 -0
- package/core/.rule/general/pipeline.yaml +246 -0
- package/core/.rule/general/principles.yaml +77 -0
- package/core/.rule/general/project-self-containment.yaml +253 -0
- package/core/.rule/general/rust-verify-target.yaml +80 -0
- package/core/.rule/general/specialist-identity-contract.yaml +153 -0
- package/core/.rule/general/sunny-access-control.yaml +60 -0
- package/core/.rule/general/synapse-code-intelligence.yaml +375 -0
- package/core/.rule/general/synapse-cognitive-fusion.yaml +301 -0
- package/core/.rule/general/system-invariants.yaml +248 -0
- package/core/.rule/general/task-pool-routing.yaml +102 -0
- package/core/.rule/general/worktree-isolation.yaml +79 -0
- package/core/.rule/pipeline/PPL-002-completed-task-archival.yaml +528 -0
- package/core/.rule/pipeline/auto-update.yaml +282 -0
- package/core/.rule/pipeline/task-execution.yaml +368 -0
- package/core/.rule/tools/check-permissions.js +197 -0
- package/core/.specialist/Amelia/README.md +329 -0
- package/core/.specialist/Amelia/capability.yaml +386 -0
- package/core/.specialist/Amelia/knowledge.yaml +217 -0
- package/core/.specialist/Amelia/principles.yaml +10 -0
- package/core/.specialist/Amelia/rule.yaml +79 -0
- package/core/.specialist/Amelia/workflow.yaml +241 -0
- package/core/.specialist/Ansuz/HEURISTIC_BOOT_ROUTING.md +39 -0
- package/core/.specialist/Ansuz/ROUTING.md +35 -0
- package/core/.specialist/Ansuz/RULES.md +79 -0
- package/core/.specialist/Ansuz/SYSTEM_DOC.md +611 -0
- package/core/.specialist/Ansuz/VERIFY.md +38 -0
- package/core/.specialist/Ansuz/capability.yaml +338 -0
- package/core/.specialist/Ansuz/cognitive_principles.yaml +62 -0
- package/core/.specialist/Ansuz/concepts/agent_types.yaml +22 -0
- package/core/.specialist/Ansuz/concepts/autonomous_agenda.yaml +170 -0
- package/core/.specialist/Ansuz/concepts/autonomy_contract.yaml +126 -0
- package/core/.specialist/Ansuz/concepts/boot_ritual.yaml +107 -0
- package/core/.specialist/Ansuz/concepts/boot_rules_summary.yaml +16 -0
- package/core/.specialist/Ansuz/concepts/capability_pointers.yaml +21 -0
- package/core/.specialist/Ansuz/concepts/cognitive_fluency_contract.yaml +264 -0
- package/core/.specialist/Ansuz/concepts/cognitive_systems_competence_map.yaml +14 -0
- package/core/.specialist/Ansuz/concepts/cognitive_systems_hippocampus.yaml +14 -0
- package/core/.specialist/Ansuz/concepts/cognitive_systems_overview.yaml +27 -0
- package/core/.specialist/Ansuz/concepts/cognitive_systems_serendipity.yaml +14 -0
- package/core/.specialist/Ansuz/concepts/cognitive_systems_turn_context.yaml +13 -0
- package/core/.specialist/Ansuz/concepts/commitment_layer.yaml +150 -0
- package/core/.specialist/Ansuz/concepts/completion_tracking.yaml +29 -0
- package/core/.specialist/Ansuz/concepts/deep_knowledge_pointers.yaml +28 -0
- package/core/.specialist/Ansuz/concepts/evolvable_guardrails.yaml +20 -0
- package/core/.specialist/Ansuz/concepts/fatigue_handoff.yaml +84 -0
- package/core/.specialist/Ansuz/concepts/formation_quick_ref.yaml +28 -0
- package/core/.specialist/Ansuz/concepts/guardrails_immutable.yaml +31 -0
- package/core/.specialist/Ansuz/concepts/identity.yaml +48 -0
- package/core/.specialist/Ansuz/concepts/interaction_protocol.yaml +42 -0
- package/core/.specialist/Ansuz/concepts/meta_evolvable_tier.yaml +28 -0
- package/core/.specialist/Ansuz/concepts/mindset.yaml +18 -0
- package/core/.specialist/Ansuz/concepts/phase9_verification.yaml +21 -0
- package/core/.specialist/Ansuz/concepts/phase_protocol.yaml +66 -0
- package/core/.specialist/Ansuz/concepts/principles.yaml +65 -0
- package/core/.specialist/Ansuz/concepts/self_evolution_overview.yaml +53 -0
- package/core/.specialist/Ansuz/concepts/task_first_protocol.yaml +7 -0
- package/core/.specialist/Ansuz/concepts/task_pool.yaml +119 -0
- package/core/.specialist/Ansuz/concepts/task_system.yaml +22 -0
- package/core/.specialist/Ansuz/concepts/write_verification.yaml +26 -0
- package/core/.specialist/Ansuz/core.yaml +224 -0
- package/core/.specialist/Ansuz/decisions/ADR-requirement-graph-layer.md +390 -0
- package/core/.specialist/Ansuz/elixir.yaml +70 -0
- package/core/.specialist/Ansuz/exo_cortex_tools.yaml +13 -0
- package/core/.specialist/Ansuz/experiments/boot_ablation_replay.js +197 -0
- package/core/.specialist/Ansuz/guards/blind_spot_guards.yaml +28 -0
- package/core/.specialist/Ansuz/guards/lesson_guards.yaml +203 -0
- package/core/.specialist/Ansuz/identity_contract.yaml +375 -0
- package/core/.specialist/Ansuz/knowledge.yaml +284 -0
- package/core/.specialist/Ansuz/modules/amelia_review.yaml +112 -0
- package/core/.specialist/Ansuz/modules/ddx.yaml +47 -0
- package/core/.specialist/Ansuz/modules/design_checks.yaml +47 -0
- package/core/.specialist/Ansuz/modules/identity_verification.yaml +119 -0
- package/core/.specialist/Ansuz/modules/loop_guard.yaml +36 -0
- package/core/.specialist/Ansuz/modules/prevention.yaml +84 -0
- package/core/.specialist/Ansuz/modules/prophecy.yaml +95 -0
- package/core/.specialist/Ansuz/modules/pulse_empty.yaml +36 -0
- package/core/.specialist/Ansuz/modules/qa_intake.yaml +29 -0
- package/core/.specialist/Ansuz/modules/recall.yaml +45 -0
- package/core/.specialist/Ansuz/principles.yaml +31 -0
- package/core/.specialist/Ansuz/research/EMBER_architecture_comparison.md +142 -0
- package/core/.specialist/Ansuz/rule.yaml +97 -0
- package/core/.specialist/Ansuz/self_model_seed.backup.json +100 -0
- package/core/.specialist/Ansuz/self_model_seed.json +100 -0
- package/core/.specialist/Ansuz/self_model_seed.sha256 +1 -0
- package/core/.specialist/Ansuz/tools/ansuz-handoff-snapshot.js +626 -0
- package/core/.specialist/Ansuz/tools/ansuz-handoff.js +437 -0
- package/core/.specialist/Ansuz/tools/ansuz-sdk-runner.js +1254 -0
- package/core/.specialist/Ansuz/tools/ansuz-sdk-runner.test.js +137 -0
- package/core/.specialist/Ansuz/tools/ansuz-state.js +288 -0
- package/core/.specialist/Ansuz/tools/bootstrap-anchors.js +289 -0
- package/core/.specialist/Ansuz/tools/handoff-freshness-milestone.test.js +73 -0
- package/core/.specialist/Ansuz/tools/handoff-resolver.js +207 -0
- package/core/.specialist/Ansuz/tools/handoff-session-identity.test.js +333 -0
- package/core/.specialist/Ansuz/tools/handoff-snapshot-args.test.js +25 -0
- package/core/.specialist/Ansuz/tools/install-claude-hooks.js +354 -0
- package/core/.specialist/Ansuz/tools/install-claude-hooks.test.js +110 -0
- package/core/.specialist/Ansuz/tools/package.json +16 -0
- package/core/.specialist/Ansuz/tools/preservation-coordinator.js +504 -0
- package/core/.specialist/Ansuz/tools/verify-and-land.js +1012 -0
- package/core/.specialist/Ansuz/tools/verify-and-land.test.js +1099 -0
- package/core/.specialist/Ansuz/workflow.yaml +1273 -0
- package/core/.specialist/Forseti/capability.yaml +210 -0
- package/core/.specialist/Forseti/core.yaml +176 -0
- package/core/.specialist/Forseti/principles.yaml +88 -0
- package/core/.specialist/Forseti/rule.yaml +86 -0
- package/core/.specialist/Heimdall/SKILL.md +65 -0
- package/core/.specialist/Heimdall/capability.yaml +134 -0
- package/core/.specialist/Heimdall/core.yaml +275 -0
- package/core/.specialist/Heimdall/principles.yaml +66 -0
- package/core/.specialist/Heimdall/rule.yaml +68 -0
- package/core/.specialist/Huginn/capability.yaml +234 -0
- package/core/.specialist/Huginn/core.yaml +482 -0
- package/core/.specialist/Huginn/principles.yaml +57 -0
- package/core/.specialist/Huginn/rule.yaml +121 -0
- package/core/.specialist/REFLECTION_TO_CONTROL.md +347 -0
- package/core/.specialist/ansuz_session_handoff.json +96 -0
- package/core/.specialist/experiments/rg_benchmark.js +523 -0
- package/core/.specialist/experiments/rg_benchmark_results.json +27 -0
- package/core/.specialist/shared/cognitive-primer.md +15 -0
- package/core/.specialist/shared/guard_schema.json +290 -0
- package/core/.specialist/shared/principles.yaml +213 -0
- package/core/.specialist/tests/runtime-discovery/README.md +224 -0
- package/core/.specialist/tests/runtime-discovery/active-project-polling.test.js +495 -0
- package/core/.specialist/tests/runtime-discovery/port-binding.test.js +418 -0
- package/core/.specialist/tests/runtime-discovery/resolver-priority.test.js +693 -0
- package/core/.specialist/tests/runtime-discovery/resolver-skip-on-invalid.test.js +467 -0
- package/core/.specialist/tests/runtime-discovery/resolver-symmetry.test.js +679 -0
- package/core/.specialist/tests/runtime-discovery/tail-cli-project-path.test.js +231 -0
- package/core/.specialist/tools/MEMORY_RECALL_TRIGGERS.md +156 -0
- package/core/.specialist/tools/README.md +129 -0
- package/core/.specialist/tools/__tests__/freya-cli-contract.test.js +156 -0
- package/core/.specialist/tools/__tests__/task-1863-boot-lease-reuse.test.js +99 -0
- package/core/.specialist/tools/__tests__/task-1917-orphan-detection.test.js +175 -0
- package/core/.specialist/tools/__tests__/task-1977-banner-cap.test.js +129 -0
- package/core/.specialist/tools/__tests__/task-1983-lease-project-path.test.js +96 -0
- package/core/.specialist/tools/__tests__/task-2218-boot-session-id.test.js +118 -0
- package/core/.specialist/tools/__tests__/task-2222-session-identity-interleave.test.js +203 -0
- package/core/.specialist/tools/__tests__/task-2246-reap-honesty.test.js +96 -0
- package/core/.specialist/tools/__tests__/task-2254-handoff-archive-contract.test.js +60 -0
- package/core/.specialist/tools/__tests__/task-2272-open-card-status-quote.test.js +26 -0
- package/core/.specialist/tools/__tests__/task-2374-usage-counter.test.js +34 -0
- package/core/.specialist/tools/ansuz-boot-fast.js +1120 -0
- package/core/.specialist/tools/ansuz-boot-fast.test.js +705 -0
- package/core/.specialist/tools/canary.js +501 -0
- package/core/.specialist/tools/context-injection.js +395 -0
- package/core/.specialist/tools/design-reasoning.js +493 -0
- package/core/.specialist/tools/discovery-fixtures.js +308 -0
- package/core/.specialist/tools/discovery-runner.js +460 -0
- package/core/.specialist/tools/failure-detector.js +1132 -0
- package/core/.specialist/tools/identity-scorer.js +714 -0
- package/core/.specialist/tools/jev-optional-context.js +107 -0
- package/core/.specialist/tools/jev-optional-context.test.js +152 -0
- package/core/.specialist/tools/lesson-guard.js +1084 -0
- package/core/.specialist/tools/lib/chart-path.js +159 -0
- package/core/.specialist/tools/lib/chart-path.test.js +171 -0
- package/core/.specialist/tools/lib/project-resolver.js +200 -0
- package/core/.specialist/tools/lib/project-resolver.test.js +241 -0
- package/core/.specialist/tools/memory-action-log.js +839 -0
- package/core/.specialist/tools/phase-gate.js +3754 -0
- package/core/.specialist/tools/phase-gate.test.js +85 -0
- package/core/.specialist/tools/phase9-policy.test.js +109 -0
- package/core/.specialist/tools/precompute-runner.js +444 -0
- package/core/.specialist/tools/prompt-files.test.js +169 -0
- package/core/.specialist/tools/replay-detectors.js +290 -0
- package/core/.specialist/tools/rule-engine.js +621 -0
- package/core/.specialist/tools/session-boot-ownership.test.js +117 -0
- package/core/.specialist/tools/session-project-state.test.js +65 -0
- package/core/.specialist/tools/session.js +1866 -0
- package/core/.specialist/tools/stream-lease.js +263 -0
- package/core/.specialist/tools/task-bridge-jev.test.js +170 -0
- package/core/.specialist/tools/task-bridge.js +765 -0
- package/core/.specialist/tools/verify-identity.js +453 -0
- package/core/.specialist/tools/wish-mandate-delivery.test.js +76 -0
- package/core/.system/README.md +555 -0
- package/core/.system/agent-skills/README.md +724 -0
- package/core/.system/archive/README.md +499 -0
- package/core/.system/assessment/README.md +645 -0
- package/core/.system/cli-tools/README.md +724 -0
- package/core/.system/collaboration/KNOWLEDGE.md +65 -0
- package/core/.system/collaboration/README.md +570 -0
- package/core/.system/cortex/MODULES.md +1046 -0
- package/core/.system/dashboard/FRONTEND.md +812 -0
- package/core/.system/dashboard/README.md +737 -0
- package/core/.system/external-skills/README.md +522 -0
- package/core/.system/failure/README.md +723 -0
- package/core/.system/formation/LEGACY_SKILLS.md +972 -0
- package/core/.system/formation/POST_COMPLETE.md +600 -0
- package/core/.system/formation/README.md +1092 -0
- package/core/.system/formation/TOOLS_REFERENCE.md +1145 -0
- package/core/.system/githooks/README.md +846 -0
- package/core/.system/kenaz/README.md +787 -0
- package/core/.system/memory/README.md +716 -0
- package/core/.system/pipeline/README.md +712 -0
- package/core/.system/rule/README.md +794 -0
- package/core/.system/specialist/README.md +669 -0
- package/core/.system/specialist/STRUCTURE.md +262 -0
- package/core/.system/sunny/README.md +476 -0
- package/core/.system/sunny/RULES.json +89 -0
- package/core/.system/sunny/knowledge/INDEX.md +37 -0
- package/core/.system/sunny/knowledge/agent-teams-api.md +0 -0
- package/core/.system/sunny/knowledge/ansuz-specialist.md +500 -0
- package/core/.system/sunny/knowledge/claude-cli-spawn.md +177 -0
- package/core/.system/sunny/knowledge/collaboration-api.md +925 -0
- package/core/.system/sunny/teammates/README.md +107 -0
- package/core/.system/sunny/teammates/how_to_use.yaml +101 -0
- package/core/.system/sunny/teammates/tools/headless/launcher.js +318 -0
- package/core/.system/sunny/teammates/tools/powershell/launcher.js +241 -0
- package/core/.system/sunny/teammates/tools/powershell/start.bat +50 -0
- package/core/.system/sunny/teammates/tools/utils/env.js +96 -0
- package/core/.system/sunny/teammates/tools/utils/format-output.js +329 -0
- package/core/.system/sunny/tools.json +101 -0
- package/core/.vendor/js-yaml.cjs +2795 -0
- package/core/.vendor/require-yaml.cjs +8 -0
- package/core/CLAUDE.md +172 -0
- package/core/KENAZ_CORE_VERSION +1 -0
- package/core/dev/ALAYA_FIELD_FEEDBACK_DESIGN.md +126 -0
- package/core/dev/ANSUZ_SCAFFOLD_DIET_2026-09-20.md +231 -0
- package/core/dev/ANSUZ_STARTUP_INJECTION_REVIEW.md +355 -0
- package/core/dev/ANSUZ_WORKFLOW_MODERNIZATION.md +143 -0
- package/core/dev/CORTEX_META_GRAPH_ASSESSMENT.md +99 -0
- package/core/dev/FREYA_DOWNSTREAM_BUGS_2026-09-19.md +44 -0
- package/core/dev/GLOBAL_MEMORY_IDENTITY_CONSOLIDATION.md +263 -0
- package/core/dev/GUARD_EFFECTIVENESS_DESIGN.md +180 -0
- package/core/dev/HEALTH_GATE_VARIANCE_ASSESSMENT.md +271 -0
- package/core/dev/HOOK_DIET_2026-09-21.md +126 -0
- package/core/dev/JEV_INTEGRATION_PLAN.md +189 -0
- package/core/dev/KENAZ_AGENT_CONNECT_v1_SPEC.md +139 -0
- package/core/dev/MOBILE_DESIGN_SPEC.md +569 -0
- package/core/dev/MULTI_ANSUZ_SPEC.md +136 -0
- package/core/dev/MUNINN_RETIREMENT.md +115 -0
- package/core/dev/PHASE3_ATTRIBUTION_POPULATION_DIAGNOSIS.md +211 -0
- package/core/dev/PHASE3_P3-2B_TASKLEVEL_PROBE.md +83 -0
- package/core/dev/PHASE3_P3-2_POPULATION_DISJOINTNESS.md +83 -0
- package/core/dev/PHASE3_P3-2_REMEASURE_REPORT.md +231 -0
- package/core/dev/PREDICTION_GROUNDING_DESIGN.md +376 -0
- package/core/dev/PRODUCT_SPEC.md +317 -0
- package/core/dev/PRODUCT_SPEC_FOR_AI_tepmlate.md +716 -0
- package/core/dev/PROJECT_REGISTRY_DUPLICATE_ASSESSMENT.md +86 -0
- package/core/dev/STARTUP_TIMELINE_TASK_1995.md +181 -0
- package/core/dev/SYNAPSE_WORKER_PER_PROJECT.md +881 -0
- package/core/dev/TASK_1680_P3-3_DISPATCH_SPEC.md +96 -0
- package/core/dev/TASK_1684_CALIBRATION_IMPACT_ANALYSIS.md +170 -0
- package/core/dev/TASK_1684_FORK_A_GOODHART_DESIGN.md +171 -0
- package/core/dev/TASK_1684_P3_2_REMEASURE_VALIDATION.md +262 -0
- package/core/dev/TASK_1903_SIGNAL_AUC_MEASUREMENT.md +207 -0
- package/core/dev/TASK_1904_BRANCH_INTEGRITY_AUDIT.md +146 -0
- package/core/dev/ansuz-eval/README.md +38 -0
- package/core/dev/ansuz-eval/build-fixtures.js +87 -0
- package/core/dev/ansuz-eval/grade.js +96 -0
- package/core/dev/ansuz-eval/grader.test.js +117 -0
- package/core/dev/ansuz-eval/tasks.md +32 -0
- package/core/dev/archive/ansuz/BOOT_FULL.md +141 -0
- package/core/dev/archive/ansuz/agents/ansuz-boot.md +176 -0
- package/core/dev/archive/ansuz/designs/alaya-passive-surfacing.md +84 -0
- package/core/dev/archive/ansuz/designs/cc_memory_indexing.md +307 -0
- package/core/dev/archive/ansuz/designs/check_guard.md +322 -0
- package/core/dev/archive/ansuz/designs/consequence_loop.md +227 -0
- package/core/dev/archive/ansuz/designs/mcp_kuzudb_lock_architecture.md +312 -0
- package/core/dev/archive/ansuz/designs/phase0_git_enrichment.md +278 -0
- package/core/dev/archive/ansuz/designs/refactor_mcp_tools_synapse.md +490 -0
- package/core/dev/archive/ansuz/designs/semantic_markdown_extraction.md +167 -0
- package/core/dev/archive/ansuz/designs/task_1372_prompt_injects_into.md +193 -0
- package/core/dev/archive/ansuz/designs/thought_stack.md +286 -0
- package/core/dev/archive/ansuz/investigations/reachability_unified_api_2026-04-11.md +330 -0
- package/core/dev/archive/ansuz/investigations/synapse_orphans_indexing_bug_2026-04-11.md +352 -0
- package/core/dev/archive/ansuz/investigations/synapse_orphans_reindex_verification_2026-04-11.md +48 -0
- package/core/dev/archive/ansuz/investigations/synapse_orphans_verification_2026-04-11.md +170 -0
- package/core/dev/archive/ansuz/legacy/kernel.yaml +399 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_0.yaml +418 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_1.yaml +376 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_10.yaml +35 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_11.yaml +582 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_2.yaml +94 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_3.yaml +274 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_4.yaml +67 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_5.yaml +32 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_6.yaml +91 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_7.yaml +58 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_8.yaml +146 -0
- package/core/dev/archive/ansuz/legacy/phases/phase_9.yaml +424 -0
- package/core/dev/archive/ansuz/specs/INDEX_bifrost_hook_parity.yaml +92 -0
- package/core/dev/archive/ansuz/specs/active-inference-formation.yaml +453 -0
- package/core/dev/archive/ansuz/specs/brief_TASK_1228_userpromptsubmit.yaml +191 -0
- package/core/dev/archive/ansuz/specs/brief_TASK_1229_pretooluse_posttooluse.yaml +250 -0
- package/core/dev/archive/ansuz/specs/brief_TASK_1230_sessionstart_stop.yaml +301 -0
- package/core/dev/archive/ansuz/specs/brief_TASK_1231_verification.yaml +272 -0
- package/core/dev/design/PHASE4_DESIGN.md +559 -0
- package/core/dev/design/dashboard_docs_integration_PLAN.md +388 -0
- package/core/dev/design/matrix_format_ABBREVIATIONS.md +655 -0
- package/core/dev/design/matrix_format_DESIGN_SPEC.md +601 -0
- package/core/dev/design/matrix_format_DISCUSSION_SUMMARY.md +484 -0
- package/core/dev/design/matrix_format_EXAMPLE_memory.matrix +13 -0
- package/core/dev/design/matrix_format_EXAMPLE_project_status.matrix +9 -0
- package/core/dev/design/matrix_format_EXAMPLE_tasks.matrix +14 -0
- package/core/dev/design/matrix_format_SYNTAX_BNF.md +754 -0
- package/core/dev/design/matrix_format_TOKEN_COMPARISON.md +484 -0
- package/core/dev/design/matrix_v2_COGNITIVE_DESIGN.md +503 -0
- package/core/dev/design/matrix_v2_DISCUSSION_SUMMARY.md +414 -0
- package/core/dev/design/matrix_v2_EXAMPLE.matrix +67 -0
- package/core/dev/designs/EXPANDABLE_CARD_DESIGN_SPEC.md +509 -0
- package/core/dev/jev-eval/ACTIVE_CORRECTION.md +52 -0
- package/core/dev/jev-eval/ANNOTATION_GUIDE.md +68 -0
- package/core/dev/jev-eval/BASELINE_COMPARISON.md +45 -0
- package/core/dev/jev-eval/CONTEXT_REVIEW.md +31 -0
- package/core/dev/jev-eval/EVALUATION_STATUS.md +39 -0
- package/core/dev/jev-eval/HISTORICAL_CORPUS.md +33 -0
- package/core/dev/jev-eval/PACKAGING_VERIFICATION.md +42 -0
- package/core/dev/jev-eval/README.md +101 -0
- package/core/dev/jev-eval/THALAMUS_RANKING.md +58 -0
- package/core/dev/jev-eval/VALIDATION_PROGRESS.md +54 -0
- package/core/dev/jev-eval/cli.mjs +50 -0
- package/core/dev/jev-eval/context.mjs +183 -0
- package/core/dev/jev-eval/corpus.mjs +186 -0
- package/core/dev/jev-eval/fixtures/conversation-review.seed.jsonl +6 -0
- package/core/dev/jev-eval/fixtures/synthetic.jsonl +11 -0
- package/core/dev/jev-eval/metrics.mjs +67 -0
- package/core/dev/jev-eval/muninn-baseline.mjs +119 -0
- package/core/dev/jev-eval/providers.mjs +95 -0
- package/core/dev/jev-eval/reports/annotation-pilot-2026-09-20.json +32 -0
- package/core/dev/jev-eval/reports/context-review-2026-09-20.json +48 -0
- package/core/dev/jev-eval/reports/conversation-review.pending.json +151 -0
- package/core/dev/jev-eval/reports/history-corpus-2026-09-20.json +52 -0
- package/core/dev/jev-eval/reports/length-probe-2026-09-20.json +133 -0
- package/core/dev/jev-eval/reports/live-smoke-2026-09-20.json +616 -0
- package/core/dev/jev-eval/reports/muninn-baseline-2026-09-20.json +52 -0
- package/core/dev/jev-eval/reports/resilience-2026-09-20.json +520 -0
- package/core/dev/jev-eval/reports/rules-baseline.json +197 -0
- package/core/dev/jev-eval/reports/synthetic-review.pending.json +357 -0
- package/core/dev/jev-eval/resilience.mjs +71 -0
- package/core/dev/jev-eval/review.mjs +78 -0
- package/core/dev/jev-eval/schema.mjs +87 -0
- package/core/dev/jev-eval/test/context.test.mjs +95 -0
- package/core/dev/jev-eval/test/corpus.test.mjs +102 -0
- package/core/dev/jev-eval/test/eval.test.mjs +146 -0
- package/core/dev/jev-eval/test/muninn-baseline.test.mjs +71 -0
- package/core/dev/jev-eval/test/resilience.test.mjs +72 -0
- package/core/dev/jev-eval/test/review.test.mjs +61 -0
- package/core/dev/pipeline/LOCK_STATE.json +4 -0
- package/core/dev/pipeline/OPTIMIZATION_PROPOSALS.json +32 -0
- package/core/dev/pipeline/PHASE3_PLANNING_SUMMARY.md +338 -0
- package/core/dev/pipeline/PIPELINE_CONFIG.json +46 -0
- package/core/dev/pipeline/PROGRESS.md +113 -0
- package/core/dev/pipeline/PROGRESS_template.md +860 -0
- package/core/dev/pipeline/QA_REPORT_F008.md +431 -0
- package/core/dev/pipeline/QA_REPORT_LOCAL_CLAUDE_DIAGNOSIS.md +583 -0
- package/core/dev/pipeline/README.md +210 -0
- package/core/dev/pipeline/SELF_ENHANCEMENT_STATUS.json.bak +47 -0
- package/core/dev/pipeline/TASK_QUEUE.json +192 -0
- package/core/dev/pipeline/WORKFLOW_GUIDE.md +824 -0
- package/core/dev/pipeline/archived/TASK_015_tauri_init.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_016_tauri_commands.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_017_rust_pty.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_018_tauri_bundle.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_019_release_workflow.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_020_update_ui.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_021_update_notification_ui.yaml +55 -0
- package/core/dev/pipeline/archived/TASK_022_remove_node_pty.yaml +65 -0
- package/core/dev/pipeline/archived/TASK_023_api_routes_migration.yaml +86 -0
- package/core/dev/pipeline/archived/TASK_024_feature_id_system.yaml +51 -0
- package/core/dev/pipeline/archived/TASK_025_executable_acceptance.yaml +66 -0
- package/core/dev/pipeline/archived/TASK_026_spec_versioning.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_027_drift_detection.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_028_tauri_pty_manager.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_029_tauri_pty_commands.yaml +62 -0
- package/core/dev/pipeline/archived/TASK_030_frontend_session_adapter.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_031_terminal_tauri_events.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_041_scheduler_data_models.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_042_path_lock_manager.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_043_process_manager.yaml +57 -0
- package/core/dev/pipeline/archived/TASK_044_git_integration.yaml +85 -0
- package/core/dev/pipeline/archived/TASK_045_task_queue_manager.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_046_scheduler_dispatcher.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_047_tauri_scheduler_commands.yaml +55 -0
- package/core/dev/pipeline/archived/TASK_048_realtime_status_push.yaml +92 -0
- package/core/dev/pipeline/archived/TASK_049_frontend_scheduler_integration.yaml +58 -0
- package/core/dev/pipeline/archived/TASK_050_error_recovery.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_051_optimization_proposals_backend.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_052_optimization_proposals_ui.yaml +58 -0
- package/core/dev/pipeline/archived/TASK_053_optimization_proposals_diff_view.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_054_multi_project_scheduler_state.yaml +76 -0
- package/core/dev/pipeline/archived/TASK_055_scheduler_api_project_scope.yaml +64 -0
- package/core/dev/pipeline/archived/TASK_056_frontend_scheduler_project_scope.yaml +62 -0
- package/core/dev/pipeline/archived/TASK_057_scheduler_module_structure.yaml +54 -0
- package/core/dev/pipeline/archived/TASK_058_scheduler_models_extract.yaml +51 -0
- package/core/dev/pipeline/archived/TASK_059_scheduler_core_extract.yaml +51 -0
- package/core/dev/pipeline/archived/TASK_060_scheduler_commands_extract.yaml +51 -0
- package/core/dev/pipeline/archived/TASK_061_scheduler_specialized_commands.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_062_process_manager_module_structure.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_063_process_manager_core_extract.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_064_recovery_module_structure.yaml +57 -0
- package/core/dev/pipeline/archived/TASK_065_recovery_core_extract.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_066_self_enhancement_scheduler_models.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_067_agents_section_quick_actions.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_067_self_enhancement_scheduler_core.yaml +65 -0
- package/core/dev/pipeline/archived/TASK_068_agent_workflow_rule_enhancement.yaml +46 -0
- package/core/dev/pipeline/archived/TASK_068_self_enhancement_tauri_commands.yaml +61 -0
- package/core/dev/pipeline/archived/TASK_069_agent_wizard_responsive.yaml +46 -0
- package/core/dev/pipeline/archived/TASK_069_self_enhancement_ui.yaml +62 -0
- package/core/dev/pipeline/archived/TASK_070_preview_panel_redesign.yaml +62 -0
- package/core/dev/pipeline/archived/TASK_070_self_enhancement_integration.yaml +57 -0
- package/core/dev/pipeline/archived/TASK_071_preview_panel_qa.yaml +98 -0
- package/core/dev/pipeline/archived/TASK_071_single_agent_config.yaml +57 -0
- package/core/dev/pipeline/archived/TASK_072_preview_modal_esc_fix.yaml +71 -0
- package/core/dev/pipeline/archived/TASK_072_product_review_module.yaml +57 -0
- package/core/dev/pipeline/archived/TASK_073_agents_page_component_refactor.yaml +68 -0
- package/core/dev/pipeline/archived/TASK_073_rename_single_agent.yaml +64 -0
- package/core/dev/pipeline/archived/TASK_074_preview_mode_ui_diagnosis.yaml +71 -0
- package/core/dev/pipeline/archived/TASK_074_single_agent_ui_toggle.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_075_agents_page_component_refactor.yaml +69 -0
- package/core/dev/pipeline/archived/TASK_075_single_agent_integration_test.yaml +75 -0
- package/core/dev/pipeline/archived/TASK_076_preview_mode_ui_diagnosis.yaml +70 -0
- package/core/dev/pipeline/archived/TASK_076_single_agent_continuous_mode.yaml +65 -0
- package/core/dev/pipeline/archived/TASK_077_failure_directory_structure.yaml +51 -0
- package/core/dev/pipeline/archived/TASK_077_single_agent_typescript_types.yaml +59 -0
- package/core/dev/pipeline/archived/TASK_078_fail_001_agent_no_output.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_078_single_agent_ui_simplify.yaml +74 -0
- package/core/dev/pipeline/archived/TASK_079_dashboard_rwd_responsive.yaml +63 -0
- package/core/dev/pipeline/archived/TASK_079_fail_002_json_bom.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_080_fail_003_queue_pollution.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_080_http_server_module.yaml +55 -0
- package/core/dev/pipeline/archived/TASK_081_failure_logging_tools.yaml +57 -0
- package/core/dev/pipeline/archived/TASK_081_http_api_routes.yaml +62 -0
- package/core/dev/pipeline/archived/TASK_082_frontend_api_adapter.yaml +51 -0
- package/core/dev/pipeline/archived/TASK_082_integrate_failure_to_workflow.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_083_cleanup_nextjs_api.yaml +48 -0
- package/core/dev/pipeline/archived/TASK_083_task_queue_cleanup.yaml +67 -0
- package/core/dev/pipeline/archived/TASK_084_rust_dead_code_cleanup.yaml +66 -0
- package/core/dev/pipeline/archived/TASK_084_scheduler_sync_logic_fix.yaml +68 -0
- package/core/dev/pipeline/archived/TASK_085_http_api_web_mode_fixes.yaml +46 -0
- package/core/dev/pipeline/archived/TASK_085_memory_consolidation_0204_0205.yaml +98 -0
- package/core/dev/pipeline/archived/TASK_086_collapsible_preview_panel.yaml +161 -0
- package/core/dev/pipeline/archived/TASK_086_web_mode_feature_parity.yaml +62 -0
- package/core/dev/pipeline/archived/TASK_087_manual_memory_archival.yaml +163 -0
- package/core/dev/pipeline/archived/TASK_087_mobile_http_control.yaml +68 -0
- package/core/dev/pipeline/archived/TASK_088_fail_003_to_ppl_002_rule.yaml +63 -0
- package/core/dev/pipeline/archived/TASK_088_mobile_http_qa.yaml +121 -0
- package/core/dev/pipeline/archived/TASK_089_agent_collaboration_topic.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_089_web_mode_monitor_fix.yaml +58 -0
- package/core/dev/pipeline/archived/TASK_090_cli_detection_fix.yaml +61 -0
- package/core/dev/pipeline/archived/TASK_090_ui_ux_lessons_topic.yaml +63 -0
- package/core/dev/pipeline/archived/TASK_091_expandable_card_design.yaml +56 -0
- package/core/dev/pipeline/archived/TASK_091_move_agent_section_components.yaml +65 -0
- package/core/dev/pipeline/archived/TASK_092_expandable_card_component.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_092_refactor_agent_create_wizard.yaml +84 -0
- package/core/dev/pipeline/archived/TASK_093_taskcard_detail_integration.yaml +57 -0
- package/core/dev/pipeline/archived/TASK_093_update_barrel_exports_test.yaml +84 -0
- package/core/dev/pipeline/archived/TASK_094_ai_agent_generation_enhancement.yaml +115 -0
- package/core/dev/pipeline/archived/TASK_094_proposalcard_detail_integration.yaml +55 -0
- package/core/dev/pipeline/archived/TASK_095_preview_fixes.yaml +129 -0
- package/core/dev/pipeline/archived/TASK_095_single_agent_task_selector.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_096_agents_page_ux_improvements.yaml +163 -0
- package/core/dev/pipeline/archived/TASK_096_single_agent_status_sync.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_097_local_claude_error_logging.yaml +59 -0
- package/core/dev/pipeline/archived/TASK_097_single_agent_monitor_events.yaml +59 -0
- package/core/dev/pipeline/archived/TASK_098_product_manager_team_phase2.yaml +60 -0
- package/core/dev/pipeline/archived/TASK_098_single_agent_refactor_test.yaml +58 -0
- package/core/dev/pipeline/archived/TASK_099_backend_team_creation.yaml +258 -0
- package/core/dev/pipeline/archived/TASK_100_frontend_team_creation.yaml +61 -0
- package/core/dev/pipeline/archived/TASK_101_orchestrator_system.yaml +403 -0
- package/core/dev/pipeline/archived/TASK_102_designer_sre_teams.yaml +90 -0
- package/core/dev/pipeline/archived/TASK_103_integration_testing.yaml +86 -0
- package/core/dev/pipeline/archived/TASK_128_integration_testing.yaml +83 -0
- package/core/dev/pipeline/archived/TASK_134_cortex_v2_phase3_file_watcher.yaml +65 -0
- package/core/dev/pipeline/archived/TASK_134_team_http_api.yaml +96 -0
- package/core/dev/pipeline/archived/TASK_135_cortex_v2_phase3_git_monitor.yaml +63 -0
- package/core/dev/pipeline/archived/TASK_136_cortex_v2_phase3_change_analyzer.yaml +69 -0
- package/core/dev/pipeline/archived/TASK_137_cortex_v2_phase3_cache_invalidator.yaml +67 -0
- package/core/dev/pipeline/archived/TASK_137_team_orchestrator_integration.yaml +93 -0
- package/core/dev/pipeline/archived/TASK_138_cortex_v2_phase3_distributed_cache.yaml +68 -0
- package/core/dev/pipeline/archived/TASK_138_matrix_folder_structure.yaml +61 -0
- package/core/dev/pipeline/archived/TASK_139_cortex_v2_phase3_cache_prewarming.yaml +65 -0
- package/core/dev/pipeline/archived/TASK_146_unified_documentation.yaml +141 -0
- package/core/dev/pipeline/archived/TASK_147_pipeline_execution_rules.yaml +76 -0
- package/core/dev/pipeline/archived/TASK_148_update_claude_md.yaml +77 -0
- package/core/dev/pipeline/archived/TASK_149_fix_nextjs_mdx_serialization.yaml +63 -0
- package/core/dev/pipeline/archived/TASK_149_update_team_docs.yaml +77 -0
- package/core/dev/pipeline/archived/TASK_150_pipeline_workflow_guide.yaml +82 -0
- package/core/dev/pipeline/archived/TASK_164_fix_smart_context_switch.yaml +49 -0
- package/core/dev/pipeline/archived/TASK_165_ctags_symbol_index.yaml +53 -0
- package/core/dev/pipeline/archived/TASK_166_symbol_search_mcp_tool.yaml +48 -0
- package/core/dev/pipeline/archived/TASK_167_llm_module_summary.yaml +55 -0
- package/core/dev/pipeline/archived/TASK_168_map_reduce_context.yaml +51 -0
- package/core/dev/pipeline/archived/TASK_169_v2_integration_testing.yaml +52 -0
- package/core/dev/pipeline/backups/archive_2026-02-04T22-07-22/TASK_QUEUE_backup.json +95 -0
- package/core/dev/pipeline/docs/multi-agent-coordination.md +446 -0
- package/core/dev/pipeline/docs/overview.md +406 -0
- package/core/dev/pipeline/docs/task-management.md +412 -0
- package/core/dev/pipeline/docs/workflows.md +32 -0
- package/core/dev/pipeline/orchestrator/schemas/task_queue_schema.json +58 -0
- package/core/dev/pipeline/orchestrator/schemas/task_schema.json +139 -0
- package/core/dev/pipeline/plans/PLAN_task_cli_refactor.md +281 -0
- package/core/dev/pipeline/tasks/.gitkeep +1 -0
- package/core/dev/pipeline/test_results/TASK_TEST_PARALLEL_002_result.txt +27 -0
- package/core/dev/plans/CLI_FIRST_INSTALL_FLOW.md +128 -0
- package/core/dev/plans/FORMATION_RETIREMENT_PLAN.md +313 -0
- package/core/dev/plans/KUZU_REPLACEMENT_PLAN.md +126 -0
- package/core/dev/schemas/PRODUCT_SPEC.example.yaml +382 -0
- package/core/dev/schemas/PRODUCT_SPEC.schema.json +800 -0
- package/core/dev/schemas/README.md +517 -0
- package/core/dev/schemas/TASK.schema.json +226 -0
- package/core/dev/schemas/dataflows_ground_truth.yaml +173 -0
- package/core/dev/schemas/dataflows_spec.yaml +160 -0
- package/core/dev/schemas/extends_ground_truth.yaml +66 -0
- package/core/dev/schemas/extends_spec.yaml +121 -0
- package/core/dev/schemas/implements_ground_truth.yaml +77 -0
- package/core/dev/schemas/implements_spec.yaml +140 -0
- package/core/dev/scripts/CRASH_DIAGNOSIS.md +47 -0
- package/core/dev/scripts/README.md +142 -0
- package/core/dev/scripts/RUN_ONCE.bat +40 -0
- package/core/dev/scripts/START_DEV_AGENT.bat +63 -0
- package/core/dev/scripts/START_MULTI_AGENT.bat +80 -0
- package/core/dev/scripts/check-alaya-report-isolation.js +255 -0
- package/core/dev/scripts/check-file-size.js +268 -0
- package/core/dev/scripts/ci-feedback.js +251 -0
- package/core/dev/scripts/clean-verify-targets.js +132 -0
- package/core/dev/scripts/cleanup-queue.js +184 -0
- package/core/dev/scripts/cleanup_queue.js +86 -0
- package/core/dev/scripts/dev_agent_scheduler.ps1 +536 -0
- package/core/dev/scripts/dispatch-stats.js +384 -0
- package/core/dev/scripts/dispatch-stats.test.js +162 -0
- package/core/dev/scripts/gc-agent-worktrees.js +461 -0
- package/core/dev/scripts/gc-agent-worktrees.test.js +48 -0
- package/core/dev/scripts/hooks/README.md +148 -0
- package/core/dev/scripts/init_git.ps1 +396 -0
- package/core/dev/scripts/init_git.sh +337 -0
- package/core/dev/scripts/init_project.ps1 +851 -0
- package/core/dev/scripts/init_project.sh +477 -0
- package/core/dev/scripts/js-yaml-resolution.test.js +48 -0
- package/core/dev/scripts/kenaz-session.js +354 -0
- package/core/dev/scripts/kenaz-session.test.js +234 -0
- package/core/dev/scripts/legacy/README.md +39 -0
- package/core/dev/scripts/legacy/dev_agent_scheduler.ps1.bak +536 -0
- package/core/dev/scripts/legacy/multi_agent_scheduler.ps1.bak +1978 -0
- package/core/dev/scripts/matrix/CLI_README.md +214 -0
- package/core/dev/scripts/matrix/README.md +310 -0
- package/core/dev/scripts/matrix/package-lock.json +4266 -0
- package/core/dev/scripts/matrix/package.json +55 -0
- package/core/dev/scripts/matrix/src/abbreviations.ts +295 -0
- package/core/dev/scripts/matrix/src/cli/commands/convert.ts +211 -0
- package/core/dev/scripts/matrix/src/cli/commands/create.ts +95 -0
- package/core/dev/scripts/matrix/src/cli/commands/delete.ts +102 -0
- package/core/dev/scripts/matrix/src/cli/commands/list-tags.ts.wip +140 -0
- package/core/dev/scripts/matrix/src/cli/commands/manage-tags.ts.wip +173 -0
- package/core/dev/scripts/matrix/src/cli/commands/query.ts +261 -0
- package/core/dev/scripts/matrix/src/cli/commands/read.ts +77 -0
- package/core/dev/scripts/matrix/src/cli/commands/stats.ts.wip +168 -0
- package/core/dev/scripts/matrix/src/cli/commands/tokens.ts.wip +126 -0
- package/core/dev/scripts/matrix/src/cli/commands/update.ts +147 -0
- package/core/dev/scripts/matrix/src/cli/commands/validate.ts +275 -0
- package/core/dev/scripts/matrix/src/cli/index.ts +350 -0
- package/core/dev/scripts/matrix/src/cli/utils/file-handler.ts +72 -0
- package/core/dev/scripts/matrix/src/cli/utils/logger.ts +65 -0
- package/core/dev/scripts/matrix/src/index.ts +62 -0
- package/core/dev/scripts/matrix/src/indexer/tag-indexer.ts +165 -0
- package/core/dev/scripts/matrix/src/parser.ts +281 -0
- package/core/dev/scripts/matrix/src/serializer.ts +151 -0
- package/core/dev/scripts/matrix/src/validator.ts +301 -0
- package/core/dev/scripts/matrix/tests/abbreviations.test.ts +122 -0
- package/core/dev/scripts/matrix/tests/cli.test.ts +177 -0
- package/core/dev/scripts/matrix/tests/convert.test.ts +143 -0
- package/core/dev/scripts/matrix/tests/parser.test.ts +203 -0
- package/core/dev/scripts/matrix/tests/query.test.ts +85 -0
- package/core/dev/scripts/matrix/tests/serializer.test.ts +197 -0
- package/core/dev/scripts/matrix/tests/validate.test.ts +141 -0
- package/core/dev/scripts/matrix/tests/validator.test.ts +167 -0
- package/core/dev/scripts/matrix/tsconfig.json +20 -0
- package/core/dev/scripts/matrix/types/index.d.ts +111 -0
- package/core/dev/scripts/matrix/types/index.d.ts.map +1 -0
- package/core/dev/scripts/matrix/types/index.js +8 -0
- package/core/dev/scripts/matrix/types/index.js.map +1 -0
- package/core/dev/scripts/matrix/types/index.ts +124 -0
- package/core/dev/scripts/memory-guard-eval.js +330 -0
- package/core/dev/scripts/memory-usage-report.js +152 -0
- package/core/dev/scripts/migrate-task-dependencies.js +198 -0
- package/core/dev/scripts/multi_agent_scheduler.ps1 +2021 -0
- package/core/dev/scripts/multi_agent_scheduler.sh +1344 -0
- package/core/dev/scripts/probe/ansuz-scaffold-ablation.js +579 -0
- package/core/dev/scripts/probe/ansuz-scaffold-ablation.test.js +244 -0
- package/core/dev/scripts/probe/hippocampus-concurrency.js +247 -0
- package/core/dev/scripts/probe/hook-latency-probe.js +315 -0
- package/core/dev/scripts/probe/hook-latency-probe.test.js +105 -0
- package/core/dev/scripts/probe/prediction-signal-auc.js +483 -0
- package/core/dev/scripts/probe/prediction-signal-auc.test.js +200 -0
- package/core/dev/scripts/reset_tasks.js +99 -0
- package/core/dev/scripts/run-rust-tests.js +207 -0
- package/core/dev/scripts/run-rust-tests.test.js +154 -0
- package/core/dev/scripts/run_once.sh +47 -0
- package/core/dev/scripts/rust-verify-worktree.js +591 -0
- package/core/dev/scripts/rust-verify-worktree.test.js +574 -0
- package/core/dev/scripts/seh_probe.rs +26 -0
- package/core/dev/scripts/setup_mcp.ps1 +192 -0
- package/core/dev/scripts/setup_mcp.sh +159 -0
- package/core/dev/scripts/setup_rag.ps1 +304 -0
- package/core/dev/scripts/setup_rag.sh +230 -0
- package/core/dev/scripts/start_dev_agent.sh +137 -0
- package/core/dev/scripts/start_multi_agent.sh +41 -0
- package/core/dev/scripts/sync_task_queue.ps1 +505 -0
- package/core/dev/scripts/sync_task_queue.sh +392 -0
- package/core/dev/scripts/task1961/RUNBOOK.md +300 -0
- package/core/dev/scripts/task1961/fixtures/never-listens.server.js +54 -0
- package/core/dev/scripts/task1961/fixtures/slow-serve.server.js +134 -0
- package/core/dev/scripts/task1961/run-fixture.js +659 -0
- package/core/dev/scripts/task1961/selftest.js +247 -0
- package/core/dev/scripts/test-pty-ws-tunnel.js +229 -0
- package/core/dev/scripts/validate_pipeline.js +197 -0
- package/core/dev/scripts/verify-isbar-capture.js +100 -0
- package/core/dev/scripts/verify-target-dir.js +69 -0
- package/core/dev/scripts/verify-task-1839.js +141 -0
- package/core/dev/specs/agent-connect/PROTOCOL.md +177 -0
- package/core/dev/specs/agent-connect/SDK_GUIDE.md +915 -0
- package/core/dev/specs/agent-connect/examples/minimal-site/README.md +137 -0
- package/core/dev/specs/agent-connect/examples/minimal-site/package.json +12 -0
- package/core/dev/specs/agent-connect/examples/minimal-site/public/index.html +116 -0
- package/core/dev/specs/agent-connect/examples/minimal-site/public/install-redirect.js +73 -0
- package/core/dev/specs/agent-connect/examples/minimal-site/server.js +516 -0
- package/core/dev/specs/agent-connect/examples/minimal-site/test/smoke.test.mjs +263 -0
- package/core/dev/specs/agent-connect/history/DEFECT_REPORT_kenaz-agent-hub_2026-09-09_v0.4-empty-prompt-no-first-turn.md +77 -0
- package/core/dev/specs/agent-connect/history/DEFECT_REPORT_kenaz-agent-hub_2026-09-09_v0.4-prompt-mojibake.md +90 -0
- package/core/dev/specs/agent-connect/history/FEATURE_REQUEST_kenaz-agent-hub_2026-09-07.md +61 -0
- package/core/dev/specs/agent-connect/history/FEATURE_REQUEST_kenaz-agent-hub_2026-09-08_relay-v1-conformance.md +76 -0
- package/core/dev/specs/agent-connect/history/FEATURE_REQUEST_kenaz-agent-hub_2026-09-09_v0.4-session-channel.md +87 -0
- package/core/dev/specs/agent-connect/history/FLOWPIX_REPLY_agent-hub_2026-09-08_relay-v1-answers.md +68 -0
- package/core/dev/specs/agent-connect/history/FLOWPIX_bridge_CONTRACT.md +642 -0
- package/core/dev/specs/agent-connect/history/FLOWPIX_spikes_relay_spike_relay.mjs +244 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-07.md +77 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-08_direction-change-v0.3.md +44 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-08_relay-v1-answers-ack.md +37 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-08_relay-v1-conformance.md +56 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-09_annotations-install-coldstart.md +44 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-09_empty-prompt-first-turn.md +56 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-09_prompt-mojibake-fixed.md +48 -0
- package/core/dev/specs/agent-connect/history/KENAZ_RESPONSE_agent-hub_2026-09-09_session-channel.md +57 -0
- package/core/dev/specs/agent-connect/history/PROTOCOL_v0.3_snapshot.md +379 -0
- package/core/dev/specs/agent-connect/history/README.md +47 -0
- package/core/dev/specs/peer/CODEX_INTERACTIVE_PUSH.md +65 -0
- package/core/dev/specs/peer/CODEX_PEER_PULL_SPIKE.md +54 -0
- package/core/dev/specs/peer/MODS_SPIKE.md +133 -0
- package/core/dev/specs/peer/PER_PROJECT_ADDRESSING.md +143 -0
- package/core/dev/specs/peer/SWARM_PROTOCOLS.md +256 -0
- package/core/dev/specs/peer/codex-hook-shim-peer.test.js +101 -0
- package/core/dev/specs/peer/codex-push-launcher.test.js +142 -0
- package/core/dev/specs/peer/setup-codex-hooks.test.js +22 -0
- package/core/dev/specs/peer-ansuz/DELIVERY_SPIKE.md +397 -0
- package/core/dev/specs/peer-ansuz/DESIGN.md +109 -0
- package/core/dev/specs/peer-ansuz/spike/claude_poke.js +129 -0
- package/core/dev/specs/peer-ansuz/spike/claude_session_start_hook.js +37 -0
- package/core/dev/specs/peer-ansuz/spike/claude_settings_hook.json +14 -0
- package/core/dev/specs/peer-ansuz/spike/codex_appserver_client.js +105 -0
- package/core/dev/specs/peer-ansuz/spike/codex_appserver_query.js +29 -0
- package/core/dev/specs/peer-ansuz/spike/extract_timeline.js +26 -0
- package/core/dev/specs/synapse/SYMBOL_IDENTITY.md +305 -0
- package/core/dev/spikes/A4-cypher-varlen.md +176 -0
- package/core/dev/spikes/bloat-rounds.sh +25 -0
- package/core/dev/spikes/mods/clean-claude.js +9 -0
- package/core/dev/spikes/mods/drive-probe.js +29 -0
- package/core/dev/spikes/mods/e2e-peer.js +71 -0
- package/core/dev/spikes/mods/evidence/e2e-peer-control.json +17 -0
- package/core/dev/spikes/mods/evidence/e2e-peer-mod.json +17 -0
- package/core/dev/spikes/mods/evidence/probe-006a6e65-d807-4bc1-b648-fd7d1edfe0b4.jsonl +4 -0
- package/core/dev/spikes/mods/evidence/probe-cebdf1f7-f9c4-47a8-83cd-7c0786177404.jsonl +12 -0
- package/core/dev/spikes/mods/evidence/probe-eb21147d-10ce-4c32-93ed-37ca37d88d5a.jsonl +11 -0
- package/core/dev/spikes/mods/evidence/stream-cebdf1f7.jsonl +18 -0
- package/core/dev/spikes/mods/kenaz-mod-probe/.claude-plugin/plugin.json +1 -0
- package/core/dev/spikes/mods/kenaz-mod-probe/hooks/hooks.json +1 -0
- package/core/dev/spikes/mods/kenaz-mod-probe/hooks/register.ts +65 -0
- package/core/dev/spikes/mods/kenaz-mod-probe/tsconfig.json +3 -0
- package/core/k-cli/README.md +423 -0
- package/core/k-cli/package.json +21 -0
- package/core/k-cli/templates/k-cli.md +91 -0
- package/core/k-cli/test/TEST_SPEC.md +134 -0
- package/core/k-cli/test/run.sh +469 -0
- package/core/plugins/kenaz/.claude-plugin/plugin.json +1 -0
- package/core/plugins/kenaz/.mcp.json +5 -0
- package/core/plugins/kenaz/INVENTORY.md +48 -0
- package/core/plugins/kenaz/README.md +55 -0
- package/core/plugins/kenaz/commands/Amelia.md +5 -0
- package/core/plugins/kenaz/commands/Ansuz.md +38 -0
- package/core/plugins/kenaz/commands/architect.md +24 -0
- package/core/plugins/kenaz/commands/incident.md +15 -0
- package/core/plugins/kenaz/commands/kenaz.md +12 -0
- package/core/plugins/kenaz/commands/review.md +14 -0
- package/core/plugins/kenaz/commands/security.md +15 -0
- package/core/plugins/kenaz/commands/spec.md +22 -0
- package/core/plugins/kenaz/commands/strategy.md +141 -0
- package/core/plugins/kenaz/commands//347/224/242/345/223/201.md +15 -0
- package/core/plugins/kenaz/commands//350/250/255/350/250/210.md +14 -0
- package/core/plugins/kenaz/commands//350/251/225/344/274/260.md +16 -0
- package/core/plugins/kenaz/guidance.md +5 -0
- package/core/plugins/kenaz/hooks/core-resolve.js +230 -0
- package/core/plugins/kenaz/hooks/hooks.json +31 -0
- package/core/plugins/kenaz/hooks/kenaz-hook.js +112 -0
- package/core/plugins/kenaz/hooks/mode.js +39 -0
- package/core/plugins/kenaz/scripts/kenaz-mode.js +46 -0
- package/core/plugins/kenaz/scripts/kenaz.test.js +123 -0
- package/core/plugins/kenaz/scripts/sync.js +45 -0
- package/core/plugins/kenaz/scripts/ygg.js +157 -0
- package/core/plugins/kenaz/skills/adapt/SKILL.md +199 -0
- package/core/plugins/kenaz/skills/animate/SKILL.md +190 -0
- package/core/plugins/kenaz/skills/architect_agent/SKILL.md +408 -0
- package/core/plugins/kenaz/skills/assessment_agent/SKILL.md +289 -0
- package/core/plugins/kenaz/skills/audit/SKILL.md +129 -0
- package/core/plugins/kenaz/skills/backend_agent/SKILL.md +407 -0
- package/core/plugins/kenaz/skills/backend_engineer/SKILL.md +788 -0
- package/core/plugins/kenaz/skills/bolder/SKILL.md +132 -0
- package/core/plugins/kenaz/skills/clarify/SKILL.md +180 -0
- package/core/plugins/kenaz/skills/code_reviewer/SKILL.md +305 -0
- package/core/plugins/kenaz/skills/code_reviewer/patterns.yaml +85 -0
- package/core/plugins/kenaz/skills/colorize/SKILL.md +158 -0
- package/core/plugins/kenaz/skills/critique/SKILL.md +118 -0
- package/core/plugins/kenaz/skills/delight/SKILL.md +317 -0
- package/core/plugins/kenaz/skills/designer/SKILL.md +845 -0
- package/core/plugins/kenaz/skills/dev_agent/SKILL.md +503 -0
- package/core/plugins/kenaz/skills/developer/SKILL.md +792 -0
- package/core/plugins/kenaz/skills/distill/SKILL.md +137 -0
- package/core/plugins/kenaz/skills/extract/SKILL.md +95 -0
- package/core/plugins/kenaz/skills/feature_agent/SKILL.md +303 -0
- package/core/plugins/kenaz/skills/frontend-design/SKILL.md +127 -0
- package/core/plugins/kenaz/skills/frontend-design/reference/color-and-contrast.md +132 -0
- package/core/plugins/kenaz/skills/frontend-design/reference/interaction-design.md +123 -0
- package/core/plugins/kenaz/skills/frontend-design/reference/motion-design.md +99 -0
- package/core/plugins/kenaz/skills/frontend-design/reference/responsive-design.md +114 -0
- package/core/plugins/kenaz/skills/frontend-design/reference/spatial-design.md +100 -0
- package/core/plugins/kenaz/skills/frontend-design/reference/typography.md +131 -0
- package/core/plugins/kenaz/skills/frontend-design/reference/ux-writing.md +107 -0
- package/core/plugins/kenaz/skills/frontend_agent/SKILL.md +327 -0
- package/core/plugins/kenaz/skills/harden/SKILL.md +358 -0
- package/core/plugins/kenaz/skills/incident_response/SKILL.md +315 -0
- package/core/plugins/kenaz/skills/normalize/SKILL.md +67 -0
- package/core/plugins/kenaz/skills/onboard/SKILL.md +243 -0
- package/core/plugins/kenaz/skills/optimize/SKILL.md +269 -0
- package/core/plugins/kenaz/skills/polish/SKILL.md +202 -0
- package/core/plugins/kenaz/skills/product_manager/SKILL.md +1091 -0
- package/core/plugins/kenaz/skills/qa_engineer/SKILL.md +1269 -0
- package/core/plugins/kenaz/skills/quieter/SKILL.md +118 -0
- package/core/plugins/kenaz/skills/security_agent/SKILL.md +312 -0
- package/core/plugins/kenaz/skills/spec_optimizer/SKILL.md +299 -0
- package/core/plugins/kenaz/skills/sre_agent/SKILL.md +1347 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/SKILL.md +377 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/charts.csv +26 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/colors.csv +97 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/icons.csv +101 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/landing.csv +31 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/products.csv +97 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/react-performance.csv +45 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/astro.csv +54 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/flutter.csv +53 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/html-tailwind.csv +56 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/jetpack-compose.csv +53 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/nextjs.csv +53 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/nuxt-ui.csv +51 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/nuxtjs.csv +59 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/react-native.csv +52 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/react.csv +54 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/shadcn.csv +61 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/svelte.csv +54 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/swiftui.csv +51 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/stacks/vue.csv +50 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/styles.csv +68 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/typography.csv +58 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/ui-reasoning.csv +101 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/ux-guidelines.csv +100 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/data/web-interface.csv +31 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/scripts/core.py +253 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/scripts/design_system.py +1067 -0
- package/core/plugins/kenaz/skills/ui-ux-pro-max/scripts/search.py +106 -0
- package/core/plugins/kenaz-peer/.claude-plugin/plugin.json +1 -0
- package/core/plugins/kenaz-peer/hooks/hooks.json +1 -0
- package/core/plugins/kenaz-peer/hooks/register.test.ts +62 -0
- package/core/plugins/kenaz-peer/hooks/register.ts +146 -0
- package/core/plugins/kenaz-peer/tsconfig.json +3 -0
- package/lib/cli.js +251 -0
- package/lib/cli.test.js +59 -0
- package/lib/dashboard.js +80 -0
- package/lib/dashboard.test.js +59 -0
- package/package.json +28 -4
- package/vendor/install-kenaz-cli.js +268 -0
- package/README.md +0 -3
|
@@ -0,0 +1,915 @@
|
|
|
1
|
+
# Kenaz Agent Connect — SDK 開發指南 (SDK_GUIDE.md)
|
|
2
|
+
|
|
3
|
+
給生態系網站開發者的接入文件。讀完本檔 + 跑過 `examples/minimal-site/`,你
|
|
4
|
+
應該能在 30 分鐘內讓自己的網站完成一次「連接 Kenaz → 審批 → 終端分頁 →
|
|
5
|
+
你的網站收到 `kenaz_hello`」的完整流程。
|
|
6
|
+
|
|
7
|
+
- 協定權威規格:`./PROTOCOL.md`(v0.4)——本檔每一個參數、錯誤訊息、行為都
|
|
8
|
+
對照該檔與 Kenaz 目前 main 上的實作,不臆測未實作的功能。
|
|
9
|
+
- 可跑範例:`./examples/minimal-site/`(零 npm 依賴的 Node 專案,`node
|
|
10
|
+
server.js` 三行指令起來)。
|
|
11
|
+
- 適用範圍:**Kenaz 自家生態系網站**(目前是 app.kenazai.com、FlowPix),
|
|
12
|
+
不對外公開推廣。
|
|
13
|
+
|
|
14
|
+
你不需要知道 Kenaz 內部怎麼運作——只需要知道深連結長什麼樣、你的 MCP 端點
|
|
15
|
+
要滿足什麼約定,以及使用者會看到什麼畫面。
|
|
16
|
+
|
|
17
|
+
---
|
|
18
|
+
|
|
19
|
+
## 0. 心智模型
|
|
20
|
+
|
|
21
|
+
```
|
|
22
|
+
你的網站(瀏覽器) 使用者本機的 Kenaz 你的網站(後端 MCP)
|
|
23
|
+
│ │ │
|
|
24
|
+
│ 1. 按「連接」 │ │
|
|
25
|
+
│──── kenaz://connect?... ─────>│ │
|
|
26
|
+
│ (OS 開啟 Kenaz) │ │
|
|
27
|
+
│ │ 2. 顯示審批卡(第 3 節) │
|
|
28
|
+
│ │ 使用者按下「允許」 │
|
|
29
|
+
│ │ 3. 開一個終端分頁, │
|
|
30
|
+
│ │ agent 掛著你的 MCP │
|
|
31
|
+
│ │────── kenaz_hello ──────────>│
|
|
32
|
+
│ │ │ 4. 記錄「已連線」
|
|
33
|
+
│<─────────(你自己的通道,例如輪詢/SSE)───────────────────────│
|
|
34
|
+
│ 5. UI 顯示「已連線」 │
|
|
35
|
+
```
|
|
36
|
+
|
|
37
|
+
三個事實決定了整份文件的形狀:
|
|
38
|
+
|
|
39
|
+
1. **深連結本身就是完整的請求。** 沒有事先的配對步驟,沒有 relay,沒有
|
|
40
|
+
WebSocket。使用者按下「連接」那一下(同意深連結)就是 launch 本身。
|
|
41
|
+
2. **Kenaz 沒有回程通道。** 你的網站完全靠 agent 呼叫你自己 MCP 的
|
|
42
|
+
`kenaz_hello` 工具才知道「連上了」。深連結送出去之後,Kenaz 不會再告訴
|
|
43
|
+
你任何事——不管是審批被拒絕、使用者沒裝 Kenaz、還是 agent 執行失敗。
|
|
44
|
+
3. **Bearer 驗證是你的責任,不是 Kenaz 的。** Kenaz 只是把深連結裡的
|
|
45
|
+
`token` 原樣轉交給 agent 當 `Authorization: Bearer <token>`。誰簽發、
|
|
46
|
+
誰驗證、TTL 多長、怎麼撤銷,全部由你決定。
|
|
47
|
+
|
|
48
|
+
---
|
|
49
|
+
|
|
50
|
+
## 1. 10 分鐘接入
|
|
51
|
+
|
|
52
|
+
### 步驟一:簽 token
|
|
53
|
+
|
|
54
|
+
你的後端在使用者按下「連接」時,簽發一個一次性 bearer token(不要重複使用
|
|
55
|
+
上一次連接時發過的 token),綁定你自己的用戶 session:
|
|
56
|
+
|
|
57
|
+
```js
|
|
58
|
+
import crypto from 'node:crypto';
|
|
59
|
+
const token = crypto.randomBytes(24).toString('base64url'); // 或任何你熟悉的簽發方式
|
|
60
|
+
// 存起來:token -> { userId, createdAt, ttl },用你自己的 session 系統
|
|
61
|
+
```
|
|
62
|
+
|
|
63
|
+
Token 的唯一硬性限制(deep_link.rs 會驗證,見第 2 節):長度 ≤ 512 字元,
|
|
64
|
+
字元集限 `[A-Za-z0-9._-]`。除此之外由你自己決定形狀——JWT、隨機字串、資料
|
|
65
|
+
庫查得到的 opaque ID 都可以,只要你自己的 MCP 端點認得它。
|
|
66
|
+
|
|
67
|
+
### 步驟二:組深連結
|
|
68
|
+
|
|
69
|
+
```js
|
|
70
|
+
const params = new URLSearchParams({
|
|
71
|
+
site: 'https://your-site.example', // 你的網站 origin
|
|
72
|
+
mcp: 'https://your-site.example/mcp', // 你的 MCP 端點(完整 URL)
|
|
73
|
+
token,
|
|
74
|
+
});
|
|
75
|
+
const deepLink = `kenaz://connect?${params.toString()}`;
|
|
76
|
+
```
|
|
77
|
+
|
|
78
|
+
前端拿到這個 URL 後直接導覽過去(`window.location.href = deepLink` 或一個
|
|
79
|
+
`<a href>`):瀏覽器會把 `kenaz://` 交給 OS,OS 交給 Kenaz(若已安裝)。
|
|
80
|
+
|
|
81
|
+
### 步驟三:MCP 加 bearer 驗證與 `kenaz_hello`
|
|
82
|
+
|
|
83
|
+
你的 MCP 端點(任何符合 MCP Streamable HTTP 的伺服器都可以)必須:
|
|
84
|
+
|
|
85
|
+
```js
|
|
86
|
+
// 偽代碼——完整可跑版本見 examples/minimal-site/server.js
|
|
87
|
+
function handleMcpRequest(req, res) {
|
|
88
|
+
const auth = req.headers['authorization']; // "Bearer <token>"
|
|
89
|
+
if (!isValidToken(auth)) { res.writeHead(401); res.end(); return; }
|
|
90
|
+
// ...JSON-RPC dispatch,至少要有 kenaz_hello 工具(見第 3 節)
|
|
91
|
+
}
|
|
92
|
+
```
|
|
93
|
+
|
|
94
|
+
做完這三步,你的網站就符合 Kenaz Agent Connect 的最小約定了。完整範例見
|
|
95
|
+
`examples/minimal-site/`。
|
|
96
|
+
|
|
97
|
+
---
|
|
98
|
+
|
|
99
|
+
## 2. 深連結
|
|
100
|
+
|
|
101
|
+
### 格式
|
|
102
|
+
|
|
103
|
+
```
|
|
104
|
+
kenaz://connect?site=<url-encoded https origin>&mcp=<url-encoded https MCP url>&token=<one-time token>[&prompt=<url-encoded text>][&agent=claude|codex][&v=1]
|
|
105
|
+
```
|
|
106
|
+
|
|
107
|
+
### 參數表
|
|
108
|
+
|
|
109
|
+
| 參數 | 必填 | 說明 |
|
|
110
|
+
|---|---|---|
|
|
111
|
+
| `site` | 是 | 你的網站 origin。正式環境必須是 `https://...`。開發模式(見第 5 節)允許 `http://localhost` / `http://127.0.0.1` / `http://[::1]`。Kenaz 用 `Url::origin()` 正規化,去掉 path/query/fragment。 |
|
|
112
|
+
| `mcp` | 是 | 你的 MCP 端點,**完整 URL(含 path)**。scheme 規則與 `site` 相同。**審批卡上顯示的主要識別是這個 URL 的 host**,不是 `site`——沒有 relay 驗證過的 origin,`mcp` 是 agent 真正會連的對象。 |
|
|
113
|
+
| `token` | 是 | 一次性 bearer,≤512 字元,字元集限 `[A-Za-z0-9._-]`。Kenaz 不解讀、不落盤——只在這次啟動期間以環境變數形式交給 agent 當 MCP 的 `Authorization: Bearer` 值。 |
|
|
114
|
+
| `prompt` | 否 | 起始 prompt,交給 agent 當第一則訊息。**省略時由 Kenaz 保證仍有第一個 turn**——Kenaz 自己補一句開場訊息觸發它,你不必也不應該為了讓 agent 動起來而塞 `prompt=`(見第 9 節)。 |
|
|
115
|
+
| `agent` | 否 | 指定 CLI,`"claude"` 或 `"codex"`。省略時依使用者自己的預設設定;顯式指定但使用者沒裝該 CLI 時 launch 直接失敗,**不會靜默退回別的 CLI**。 |
|
|
116
|
+
| `v` | 否 | 協定版本,缺省 `"1"`。目前只有 `"1"` 是合法值;其他值一律 toast 拒絕,不出審批卡。 |
|
|
117
|
+
|
|
118
|
+
任何其他未知參數一律靜默忽略,不拒絕;只有格式不合法(scheme 錯、`token`
|
|
119
|
+
過長、`v` 不支援等)才拒絕。
|
|
120
|
+
|
|
121
|
+
### URL 編碼規則
|
|
122
|
+
|
|
123
|
+
每個參數值都要 `encodeURIComponent`(JS 的 `URLSearchParams` 會自動做這件
|
|
124
|
+
事——見第 1 節範例)。`site`/`mcp` 是完整 URL,所以本身就含有會被
|
|
125
|
+
percent-encode 的字元(`:`、`/`),這是預期行為,不是 bug。
|
|
126
|
+
|
|
127
|
+
### 長度限制與「大內容放哪裡」
|
|
128
|
+
|
|
129
|
+
深連結整體長度在 Windows 上約有 **2 KB** 限制(作業系統層級,不是 Kenaz 定
|
|
130
|
+
的)。這代表:
|
|
131
|
+
|
|
132
|
+
- `prompt` 只放**簡短**的起始指示。
|
|
133
|
+
- 任何長內容(畫布狀態、文件全文、使用者的完整輸入)**不要塞進
|
|
134
|
+
`prompt=`**——讓 agent 呼叫你自己 MCP 的其他工具去讀取,你的工具回傳內容
|
|
135
|
+
或一個可再次 fetch 的 URL 即可。這也是第 3 節「大檔案走 URL」的同一個
|
|
136
|
+
原則:深連結和 MCP 工具的回應都不是傳大檔案的地方,讓 agent 自己再發一
|
|
137
|
+
次請求去拿。
|
|
138
|
+
- `prompt` 支援任意 UTF-8(含 CJK),你只要照正常的
|
|
139
|
+
`encodeURIComponent` percent-encode 即可;Windows 上把內嵌 prompt 正確
|
|
140
|
+
解碼(不會被系統 ANSI codepage 亂碼化)是 Kenaz 這一側的責任,不需要你
|
|
141
|
+
額外處理。
|
|
142
|
+
|
|
143
|
+
### `v` 版本
|
|
144
|
+
|
|
145
|
+
省略時視為 `"1"`。目前整個協定只有版本 `"1"`;`v=999` 這種未來/錯誤的版本
|
|
146
|
+
號會被**明確拒絕**(toast「protocol version unsupported: 999」),不會被
|
|
147
|
+
當成 `"1"` 靜默處理,也不會出審批卡。這是刻意設計:深連結本身就是唯一的請
|
|
148
|
+
求,沒有線上協商步驟能事後補救版本不合的情況,所以必須在這裡就失敗大聲。
|
|
149
|
+
|
|
150
|
+
### `agent` 的語意
|
|
151
|
+
|
|
152
|
+
- 省略:用使用者自己在 Kenaz 裡設定的預設 CLI/model。
|
|
153
|
+
- 指定 `"claude"` 或 `"codex"`:**只有這兩個字面值**合法(但注意:`agent=`
|
|
154
|
+
不合法值目前不會被 deep_link.rs 拒絕——它會原樣往後傳,直到
|
|
155
|
+
`launch::execute` 內部解析 CLI 時才會因為「這個 CLI 不存在/未安裝」而
|
|
156
|
+
失敗,見第 6 節疑難排解表)。
|
|
157
|
+
- 指定了但使用者沒裝該 CLI:**launch 直接失敗**,不會退回使用者的預設
|
|
158
|
+
CLI——設計上你的 `agent=` 指定要不就完全生效,要不就明確失敗,不會有「我
|
|
159
|
+
以為在跑 codex 結果其實在跑 claude」這種混淆。
|
|
160
|
+
|
|
161
|
+
---
|
|
162
|
+
|
|
163
|
+
## 3. 網站 MCP 最小約定
|
|
164
|
+
|
|
165
|
+
任何網站只要提供一個符合下列約定的 MCP 端點,就能被 Kenaz 的深連結串接。
|
|
166
|
+
Kenaz **不列舉、不審核、不限制**你的 MCP 提供什麼工具——這一節只講「必須
|
|
167
|
+
有」和「建議有」的最小子集。
|
|
168
|
+
|
|
169
|
+
### 3.1 Bearer 驗證:簽發、TTL、撤銷建議
|
|
170
|
+
|
|
171
|
+
- 每個請求的 `Authorization: Bearer <token>` 必須被你的伺服器驗證是**自己
|
|
172
|
+
簽發、未過期、未撤銷**的。Kenaz 完全不驗證這個 token——它只是深連結裡
|
|
173
|
+
那串字元的原樣轉交者。這是你的網站唯一需要信任的邊界。
|
|
174
|
+
- **簽發**:每次使用者按下「連接」都簽一個新的 token,不要重複使用上一次
|
|
175
|
+
的。把 token 綁定到你自己的登入 session,這樣即使 token 外洩,攻擊者能
|
|
176
|
+
做的事也被限制在那個使用者的權限範圍內。
|
|
177
|
+
- **TTL**:建議短——這個 token 只需要活到這次終端分頁的session 期間。過期
|
|
178
|
+
就讓後續的 MCP 呼叫全部 401。
|
|
179
|
+
- **撤銷**:如果你的網站有「登出」或「撤銷授權」功能,記得同時讓對應的
|
|
180
|
+
token 立刻失效,不要只靠 TTL 到期。
|
|
181
|
+
|
|
182
|
+
### 3.1b `X-Kenaz-Session` header(選讀)與工具權限(TASK_1947)
|
|
183
|
+
|
|
184
|
+
- 除了 `Authorization: Bearer <token>`,Kenaz 對每個請求還會**一律**多送一
|
|
185
|
+
個 `X-Kenaz-Session: <launch_id>` header。`<launch_id>` 是 Kenaz 在這次
|
|
186
|
+
啟動一開始就產生的 uuid v4,同一次深連結啟動的存續期間固定不變。
|
|
187
|
+
- 這個 header 是**選讀**的——bearer 仍是唯一你必須驗證的邊界(見 3.1)。
|
|
188
|
+
它存在的用途:如果你想分辨「這次 MCP 呼叫屬於哪一次連接」,`launch_id`
|
|
189
|
+
比單純用 token 對應更直接——同一個 token 理論上可能被同一使用者的多個
|
|
190
|
+
分頁/連接重用,但 `launch_id` 保證每次啟動獨一無二。
|
|
191
|
+
- 範例(claude 分支預配的 MCP config JSON,見
|
|
192
|
+
`dev/specs/agent-connect/PROTOCOL.md` §2 fixture
|
|
193
|
+
`mcp.config_with_session_header`):
|
|
194
|
+
|
|
195
|
+
```json
|
|
196
|
+
{"mcpServers": {"flowpix": {"type": "http", "url": "https://flowpix.app/mcp", "headers": {"Authorization": "Bearer a1b2c3d4e5f60718293a4b5c6d7e8f90", "X-Kenaz-Session": "1b0d6b9a-8f2a-4c56-8abc-9f9e9c9d9e9f"}}}}
|
|
197
|
+
```
|
|
198
|
+
|
|
199
|
+
- **本機除錯**:使用者自己在本機呼叫 `GET /api/agent-hub/status` 時,每個
|
|
200
|
+
origin 底下會列出 `sessions[]`(`session_id` 就是這個 `launch_id`、還有
|
|
201
|
+
`agent`/`model`/`cwd`/`started_at`/`alive`)——這是給使用者自己排查
|
|
202
|
+
「Kenaz 現在覺得有幾個連線活著」用的本機端點,不是你的網站能呼叫的(僅
|
|
203
|
+
限 `127.0.0.1`,見第 4 節「被封鎖後怎麼辦」)。
|
|
204
|
+
- **工具權限**:使用者在審批卡按下允許,等於同意了這個 session 使用你的
|
|
205
|
+
MCP 工具——claude 分支因此會在啟動前對你的 MCP 做一次 `tools/list` 探
|
|
206
|
+
測,把**依 tool annotations 判定為安全**的工具加進
|
|
207
|
+
`--allowedTools`,讓 agent 呼叫那些工具時不用再跳一次終端機權限詢問;
|
|
208
|
+
沒被判定為安全的工具維持逐次詢問。這個允許清單**只**涵蓋你這個 MCP
|
|
209
|
+
server 的工具,不會放寬到別的工具。判定規則、探測失敗時的退回行為見
|
|
210
|
+
3.1c。codex 分支目前沒有等效的「只允許某個 MCP」設定可加、也不套用
|
|
211
|
+
annotations 分級(查證結論見 PROTOCOL.md「codex 限制」一節)——如果使
|
|
212
|
+
用者用 codex 跑你的網站,第一次工具呼叫是否會跳審批目前未完全驗證。
|
|
213
|
+
|
|
214
|
+
### 3.1c `--allowedTools` 依 tool annotations 分級(TASK_1949,Agent Connect C-1b)
|
|
215
|
+
|
|
216
|
+
- **這是什麼**:3.1b 提到 claude 分支會預先允許部分工具跳過終端機權限詢
|
|
217
|
+
問——這一節說明「部分」是怎麼判定的。MCP 規格的 tool `annotations`
|
|
218
|
+
(`readOnlyHint`、`destructiveHint`、`idempotentHint`、`openWorldHint`)
|
|
219
|
+
讓你不用發明新協定就能告訴 Kenaz「這個工具有沒有後果」。
|
|
220
|
+
- **判定規則**:你的 `tools/list` 回應裡,一個工具的
|
|
221
|
+
`annotations.readOnlyHint == true` **或** `annotations.destructiveHint ==
|
|
222
|
+
false` 時,會被加進 `--allowedTools`。**沒有 `annotations` 物件、或有
|
|
223
|
+
`annotations` 但沒寫 `destructiveHint`,一律不會被預先允許**——MCP 規
|
|
224
|
+
格對 `destructiveHint` 的預設值就是 `true`,Kenaz 沿用這個保守預設,不
|
|
225
|
+
是自創規則。範例(見 PROTOCOL.md §2 fixture
|
|
226
|
+
`mcp.tools_list_with_annotations`):
|
|
227
|
+
|
|
228
|
+
```json
|
|
229
|
+
{"jsonrpc": "2.0", "id": 1, "result": {"tools": [{"name": "kenaz_hello", "annotations": {"destructiveHint": false, "idempotentHint": true}}, {"name": "kenaz_wait_for_message", "annotations": {"readOnlyHint": true}}, {"name": "echo", "annotations": {"readOnlyHint": true}}, {"name": "delete_all_data", "annotations": {"destructiveHint": true}}]}}
|
|
230
|
+
```
|
|
231
|
+
|
|
232
|
+
這個範例裡只有前三個工具會被預先允許;`delete_all_data`
|
|
233
|
+
(`destructiveHint: true`)維持每次呼叫都跳終端機詢問。
|
|
234
|
+
- **探測時機與逾時**:Kenaz 在組出啟動腳本前,對你的 `mcp=` 端點送一次
|
|
235
|
+
`tools/list` JSON-RPC 請求(header 與其他 MCP 呼叫相同:`Authorization:
|
|
236
|
+
Bearer <token>` + `X-Kenaz-Session`),**逾時 3 秒**。
|
|
237
|
+
- **探測失敗時的退回**:逾時、HTTP 非 2xx、回應不是合法 JSON、或回應裡
|
|
238
|
+
沒有 `result.tools` 陣列——**任何一種都不會預先允許任何工具**(不是退
|
|
239
|
+
回舊版的全部允許,也不代表你的 MCP 整個不能用),並且會在啟動的終端
|
|
240
|
+
分頁第一行印出失敗原因,方便你自己的使用者(或你自己除錯時)看到
|
|
241
|
+
「為什麼每次工具呼叫都要按一下」。探測成功時同樣會印一行摘要(例如
|
|
242
|
+
`pre-allowed 2/4 MCP tools by annotations`)。
|
|
243
|
+
- **你不需要做任何事才能運作**——不標 annotations 完全不影響功能,你的
|
|
244
|
+
工具照樣能被呼叫,唯一差別是每次呼叫都會在終端機多跳一次確認。標了
|
|
245
|
+
annotations 純粹是幫使用者少按幾次。
|
|
246
|
+
|
|
247
|
+
### 3.2 `kenaz_hello`:回應形狀與逾時處理
|
|
248
|
+
|
|
249
|
+
- **建議**(不是強制,但沒有它你的網站永遠不知道有沒有連上)提供一個
|
|
250
|
+
`kenaz_hello` 工具:**無參數**,成功回:
|
|
251
|
+
|
|
252
|
+
```json
|
|
253
|
+
{"jsonrpc": "2.0", "id": 1, "result": {"content": [{"type": "text", "text": "{\"ok\":true}"}], "isError": false}}
|
|
254
|
+
```
|
|
255
|
+
|
|
256
|
+
對應的呼叫長這樣:
|
|
257
|
+
|
|
258
|
+
```json
|
|
259
|
+
{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "kenaz_hello", "arguments": {}}}
|
|
260
|
+
```
|
|
261
|
+
|
|
262
|
+
- Kenaz 注入給 agent 的人格會在處理 `prompt` **之前**呼叫這個工具一次
|
|
263
|
+
——這是你的網站得知「Kenaz 真的連上了」的**唯一方式**(沒有回程通道,
|
|
264
|
+
見第 0 節)。
|
|
265
|
+
- **逾時處理**:建議設 **60 秒**。深連結送出去之後,如果 60 秒內沒有收到
|
|
266
|
+
`kenaz_hello` 呼叫,就把前端 UI 從「連線中」切到「請安裝 Kenaz」之類的
|
|
267
|
+
安裝指引(第 4 節有完整的偵測/fallback 邏輯)。60 秒逾時涵蓋了「使用者
|
|
268
|
+
沒裝 Kenaz」「使用者拒絕了審批卡」「使用者封鎖了你的網站」這幾種你**分
|
|
269
|
+
辨不出來**的情況——見第 6 節疑難排解。
|
|
270
|
+
|
|
271
|
+
### 3.3 工具命名:`mcp__<name>__` 前綴由 Kenaz 決定
|
|
272
|
+
|
|
273
|
+
Kenaz 把你的 MCP 掛給 agent 時,會用一個從你的 **`mcp=` URL 的 hostname**
|
|
274
|
+
衍生出來的名字當作 MCP server 名稱,agent 看到的完整工具名稱因此是
|
|
275
|
+
`mcp__<name>__<你的工具名>`:
|
|
276
|
+
|
|
277
|
+
- `mcp=https://flowpix.app/mcp` → `<name>` = `flowpix` → agent 呼叫的是
|
|
278
|
+
`mcp__flowpix__kenaz_hello`
|
|
279
|
+
- `mcp=https://sub.example.com/mcp` → `<name>` = `sub`(取 hostname 第一個
|
|
280
|
+
`.` 之前的 label)
|
|
281
|
+
- 衍生規則:hostname 的第一個 label,非英數字元一律換成 `_`;解析失敗或
|
|
282
|
+
結果是空字串則用 `site` 當預設名。
|
|
283
|
+
|
|
284
|
+
**你完全不用自己決定或宣告這個前綴**——你只要決定你自己的工具名(例如
|
|
285
|
+
`kenaz_hello`、`echo`),Kenaz 會自動加上從你的 MCP hostname 衍生的
|
|
286
|
+
server 名稱前綴。這代表:你的 `mcp=` hostname 最好是穩定、看起來像正常英
|
|
287
|
+
數字元的網域名稱,不要依賴前綴本身帶有特殊語意。
|
|
288
|
+
|
|
289
|
+
### 3.4 其餘工具完全自訂
|
|
290
|
+
|
|
291
|
+
你的產品邏輯(生圖、寫回、查詢……)全部經由自訂的 MCP 工具與 agent 互動,
|
|
292
|
+
Kenaz 對這段流量完全不可見。設計這些工具時記得:
|
|
293
|
+
|
|
294
|
+
- **大檔案走 URL**:工具的回應不要塞入大量二進位或超長文字——回傳一個
|
|
295
|
+
你自己網站可以再次驗證 bearer 後提供內容的 URL,讓 agent 用你的其他機
|
|
296
|
+
制(例如你自己網站上的一個已認證下載端點)去取,而不是塞進單一次
|
|
297
|
+
JSON-RPC 回應裡。這與深連結 `prompt=` 的 2 KB 限制是同一個設計原則的兩
|
|
298
|
+
處體現。
|
|
299
|
+
- **沒有結束通知**:agent 的終端分頁何時關閉,你的網站不會被告知。用
|
|
300
|
+
token 撤銷或閒置逾時回收狀態,不要假設有「已結束」事件。
|
|
301
|
+
|
|
302
|
+
---
|
|
303
|
+
|
|
304
|
+
## 4. 使用者體驗
|
|
305
|
+
|
|
306
|
+
### 審批卡上會顯示什麼
|
|
307
|
+
|
|
308
|
+
使用者按下深連結、Kenaz 判斷這個 origin 不是 `always`/`blocked` 狀態時,
|
|
309
|
+
會跳出一張審批卡,內容固定是:
|
|
310
|
+
|
|
311
|
+
```
|
|
312
|
+
{site} → MCP {mcp_host} → {agent}/{model} · cwd {絕對路徑} · {prompt 前 120 字}
|
|
313
|
+
```
|
|
314
|
+
|
|
315
|
+
- **site**:深連結 `site=` 的 origin(你的網站自稱,未經第三方驗證)。
|
|
316
|
+
- **MCP host**:`mcp=` URL 的 hostname——**這是主要識別**,因為沒有 relay
|
|
317
|
+
驗證過的 origin,這個 host 才是 agent 真正會連上的對象。
|
|
318
|
+
- **agent/model**:實際會用的 CLI 與 model(沒指定 `agent=` 就是使用者的
|
|
319
|
+
預設值)。
|
|
320
|
+
- **cwd**:agent 會在哪個資料夾執行——**永遠是 Kenaz Dashboard 目前開啟的
|
|
321
|
+
專案,你的網站不能指定**。這一項刻意不可省略:使用者按允許前必須知道
|
|
322
|
+
這次操作會發生在哪個資料夾。
|
|
323
|
+
- **prompt 前 120 字**:你的 `prompt=`(若有)截斷到 120 個字元。
|
|
324
|
+
|
|
325
|
+
### 一次/永遠允許/拒絕/封鎖是什麼意思
|
|
326
|
+
|
|
327
|
+
Per-origin 四態,鍵值是你的 `site=` origin(不是 `mcp=`):
|
|
328
|
+
|
|
329
|
+
| 狀態 | 意義 | 下一次同一 origin 的深連結會發生什麼 |
|
|
330
|
+
|---|---|---|
|
|
331
|
+
| `unauthorized`(初次) | 從沒授權過 | 出審批卡 |
|
|
332
|
+
| `once`(使用者按「允許」) | 這次允許,下次再問 | 出審批卡 |
|
|
333
|
+
| `always`(使用者按「永遠允許」) | 跳過審批卡 | **直接執行**(仍會在本機記一筆事件) |
|
|
334
|
+
| `blocked`(使用者按「封鎖」,或手動編輯設定檔) | 完全不信任這個 origin | **直接拒絕,連 toast 都只在 Kenaz 本機顯示,你的網站完全看不到** |
|
|
335
|
+
|
|
336
|
+
一個被使用者暫停的 Agent Hub(tray「Pause Agent Hub」)會擋下**所有**深連
|
|
337
|
+
結,包含 `always` 狀態的 origin,而且同樣不出審批卡。
|
|
338
|
+
|
|
339
|
+
### 被封鎖後怎麼辦
|
|
340
|
+
|
|
341
|
+
**你的網站完全無法得知自己被封鎖了**——沒有回程通道,被封鎖的請求和「使
|
|
342
|
+
用者根本沒裝 Kenaz」在你的網站端看起來一模一樣(都是 `kenaz_hello`
|
|
343
|
+
逾時,見第 3.2 節與第 6 節)。能解除封鎖的只有使用者自己,在自己的機器
|
|
344
|
+
上,透過下面任一方式:
|
|
345
|
+
|
|
346
|
+
- Kenaz 本機的狀態 API(`http://127.0.0.1:18080`,僅限本機呼叫):
|
|
347
|
+
- `GET /api/agent-hub/status` 列出所有 origin 目前的狀態。
|
|
348
|
+
- `POST /api/agent-hub/revoke {"origin": "https://your-site.example"}`
|
|
349
|
+
會**整筆刪除**這個 origin 的紀錄(含 `blocked` 狀態),下一次深連結
|
|
350
|
+
會重新從 `unauthorized` 出審批卡。
|
|
351
|
+
- 或直接編輯 `~/.kenaz/agent_hub/sites.json`,把該 origin 的 `state` 改掉
|
|
352
|
+
(Kenaz 每次請求都重新讀檔,不吃記憶體快取,改完立即生效)。
|
|
353
|
+
|
|
354
|
+
**你的網站文件應該教使用者**:如果「連接」按了很久都沒反應,除了「還沒安
|
|
355
|
+
裝 Kenaz」之外,也可能是之前不小心按過「封鎖」——請使用者自己去確認。
|
|
356
|
+
|
|
357
|
+
### 「請安裝 Kenaz」的偵測與 fallback
|
|
358
|
+
|
|
359
|
+
因為沒有回程通道,偵測「使用者沒裝 Kenaz」的**唯一**方法是逾時:
|
|
360
|
+
|
|
361
|
+
1. 導覽到 `kenaz://connect?...`。
|
|
362
|
+
2. 同時開始輪詢(或用你自己的推播通道)等待 `kenaz_hello` 被呼叫。
|
|
363
|
+
3. 一段時間(建議跟 `kenaz_hello` 的 60 秒逾時對齊)內沒有收到,就顯示
|
|
364
|
+
「請安裝 Kenaz」的安裝指引頁面 / 下載連結。
|
|
365
|
+
|
|
366
|
+
`examples/minimal-site/public/index.html` 有這個邏輯的完整實作
|
|
367
|
+
(`pollStatus` + `HELLO_TIMEOUT_MS`),可以直接參考或複製。注意:瀏覽器對
|
|
368
|
+
未註冊的自訂 scheme 導覽通常是靜默失敗(頁面不會跳轉、不會拋錯),所以
|
|
369
|
+
**不要**指望用 `catch` 或 `beforeunload` 之類的事件偵測「有沒有裝」——用
|
|
370
|
+
逾時輪詢才是唯一可靠的做法。
|
|
371
|
+
|
|
372
|
+
---
|
|
373
|
+
|
|
374
|
+
## 5. 開發模式
|
|
375
|
+
|
|
376
|
+
正式環境的 `site=`/`mcp=` 必須是 `https://`。要在本機用
|
|
377
|
+
`http://localhost` / `http://127.0.0.1` / `http://[::1]` 測試(例如你的
|
|
378
|
+
開發伺服器還沒有 HTTPS),需要 Kenaz 端也進入開發模式:
|
|
379
|
+
|
|
380
|
+
- **`KENAZ_AGENT_HUB_DEV=1`**:設這個環境變數啟動 Kenaz,或者
|
|
381
|
+
- Kenaz 本身是 **debug build**(`cfg!(debug_assertions)`)——本機 `cargo
|
|
382
|
+
tauri dev` 跑起來的版本預設就滿足這個條件,不需要額外設環境變數。
|
|
383
|
+
|
|
384
|
+
**環境變數只在行程啟動當下讀取一次**。這代表:
|
|
385
|
+
|
|
386
|
+
- 如果 Kenaz 已經在背景執行(含縮到 tray),**改環境變數不會讓正在跑的那
|
|
387
|
+
個行程生效**——必須先從 tray 選單選「Quit」讓 Kenaz **完整退出**(不是
|
|
388
|
+
關視窗,關視窗通常只是縮到 tray),再重新啟動一次帶著
|
|
389
|
+
`KENAZ_AGENT_HUB_DEV=1` 的行程。
|
|
390
|
+
- 這是最常見的「明明設了環境變數,dev 深連結還是被拒絕」的原因——見第 6
|
|
391
|
+
節疑難排解表。
|
|
392
|
+
|
|
393
|
+
**Dev 深連結指向 `target\debug` 的行為**:Debug build 在啟動時(`.setup()`
|
|
394
|
+
內)會呼叫 `register_all()`,把 OS 的 `kenaz://` scheme 註冊指向**這次啟
|
|
395
|
+
動的那個執行檔**(通常是 `target\debug\kenaz-dashboard.exe`)。這代表:
|
|
396
|
+
|
|
397
|
+
- 如果你的機器上同時有「一般安裝的 Kenaz」和「本機 build 的 dev
|
|
398
|
+
Kenaz」,dev build 一啟動就會把 `kenaz://` 的 OS 註冊**改指向自己**,
|
|
399
|
+
之後所有 `kenaz://connect` 深連結(包含正式網站點出來的)都會打開 dev
|
|
400
|
+
版本,直到你重新安裝/啟動正式版覆蓋回去。開發時如果發現深連結打開了不
|
|
401
|
+
預期的 Kenaz 視窗,先檢查是不是這個原因。
|
|
402
|
+
|
|
403
|
+
---
|
|
404
|
+
|
|
405
|
+
## 6. 測試
|
|
406
|
+
|
|
407
|
+
### 對自己的 MCP 端點做煙霧測試(不需要 Kenaz)
|
|
408
|
+
|
|
409
|
+
你不需要真的裝 Kenaz 才能驗證你的 MCP 端點符不符合第 3 節的約定——因為
|
|
410
|
+
Kenaz 對你的 MCP 來說就是「一個會帶著正確 bearer 呼叫 `kenaz_hello`、以及
|
|
411
|
+
帶著錯誤/缺失 bearer 呼叫任何工具」的 HTTP client,這兩件事你可以直接用
|
|
412
|
+
`curl` 或任何 HTTP client 模擬:
|
|
413
|
+
|
|
414
|
+
```bash
|
|
415
|
+
# 1. 錯誤 bearer 必須 401
|
|
416
|
+
curl -i -X POST https://your-site.example/mcp \
|
|
417
|
+
-H 'content-type: application/json' \
|
|
418
|
+
-H 'authorization: Bearer wrong-token' \
|
|
419
|
+
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"kenaz_hello","arguments":{}}}'
|
|
420
|
+
|
|
421
|
+
# 2. 正確 bearer 呼叫 kenaz_hello 必須回 {"ok":true}
|
|
422
|
+
curl -i -X POST https://your-site.example/mcp \
|
|
423
|
+
-H 'content-type: application/json' \
|
|
424
|
+
-H 'authorization: Bearer <一個你自己簽發的合法 token>' \
|
|
425
|
+
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"kenaz_hello","arguments":{}}}'
|
|
426
|
+
```
|
|
427
|
+
|
|
428
|
+
`examples/minimal-site/test/smoke.test.mjs` 是這個模式的完整
|
|
429
|
+
`node --test` 版本(起一個真的 HTTP server、真的送請求、結束時關閉伺
|
|
430
|
+
服器),可以直接拿來當你自己 MCP 端點測試的樣板——把 `createServer()` 換
|
|
431
|
+
成你自己的伺服器啟動函式即可。
|
|
432
|
+
|
|
433
|
+
**目前沒有一個「假 Kenaz」npm 套件**能自動幫你送出 `kenaz://connect` 並模
|
|
434
|
+
擬 agent 呼叫 `kenaz_hello`(SDK 規格 §2.4 把它列為 C 期產出,尚未發佈)
|
|
435
|
+
——上面的 curl / smoke test 手法就是目前**唯一已驗證可行**的「不裝 Kenaz
|
|
436
|
+
也能測」方式:既然 Kenaz 對你的 MCP 來說只是「一個會呼叫工具的 HTTP
|
|
437
|
+
client」,直接模擬那個 client 的行為,比等一個尚未存在的模擬器更直接。
|
|
438
|
+
|
|
439
|
+
### Conformance suite(Kenaz 這一側的測試)
|
|
440
|
+
|
|
441
|
+
`dashboard/scripts/agent-connect-conformance/run.mjs` 是 Kenaz 專案自己
|
|
442
|
+
的 conformance suite,驗證兩件事:
|
|
443
|
+
|
|
444
|
+
1. `PROTOCOL.md` 裡的 `#### Fixture: deep_link.*` / `mcp.*` 區塊都是合法
|
|
445
|
+
JSON,且深連結 fixture 的形狀符合第 2 節的參數表。
|
|
446
|
+
2. Kenaz 自己的假 MCP(`dashboard/scripts/agent-hub-fake-mcp.mjs`)確實會
|
|
447
|
+
對錯誤 bearer 回 401、對 `kenaz_hello` 回 `{"ok":true}` 並記錄呼叫。
|
|
448
|
+
|
|
449
|
+
這個 suite **目前測的是 Kenaz 自己的假 MCP,不是可以指向任意第三方 MCP
|
|
450
|
+
URL 的通用工具**——它不是你驗證自己網站 MCP 端點的方式;你的驗證方式是上
|
|
451
|
+
面的 curl / smoke test。它們驗證的是同一件事(bearer 401 + `kenaz_hello`
|
|
452
|
+
形狀),只是分別跑在 Kenaz 這一側與你這一側。跑法:
|
|
453
|
+
|
|
454
|
+
```bash
|
|
455
|
+
node --test dashboard/scripts/agent-connect-conformance/run.mjs
|
|
456
|
+
```
|
|
457
|
+
|
|
458
|
+
### 對真 Kenaz 做一次端到端連接(需要安裝或本機跑 Kenaz)
|
|
459
|
+
|
|
460
|
+
1. 依第 5 節設定開發模式並完整重啟 Kenaz。
|
|
461
|
+
2. 跑起 `examples/minimal-site/`(`node server.js`)。
|
|
462
|
+
3. 瀏覽器打開範例頁面,按「連接 Kenaz」。
|
|
463
|
+
4. 預期看到:審批卡(site/MCP host/agent/model/cwd/prompt 都符合第 4 節格
|
|
464
|
+
式)→ 按「允許」→ Dashboard 開一個新的終端分頁 → agent 在處理 prompt
|
|
465
|
+
之前呼叫 `mcp__<name>__kenaz_hello` → 範例頁面的輪詢偵測到「已連線」。
|
|
466
|
+
|
|
467
|
+
---
|
|
468
|
+
|
|
469
|
+
## 7. 疑難排解
|
|
470
|
+
|
|
471
|
+
### Toast 訊息對照表
|
|
472
|
+
|
|
473
|
+
Kenaz 拒絕深連結時,只會在**自己本機**顯示一個 toast(你的網站看不到,
|
|
474
|
+
沒有回程通道)。下表逐字對照 Kenaz 目前 main 上
|
|
475
|
+
`dashboard/src-tauri/src/features/agent_hub/deep_link.rs` 的
|
|
476
|
+
`DeepLinkError` Display 實作與 `mod.rs::handle_incoming_url` 組出的完整
|
|
477
|
+
toast 文字:
|
|
478
|
+
|
|
479
|
+
| 情況 | 完整 toast 文字 | 你該怎麼修 |
|
|
480
|
+
|---|---|---|
|
|
481
|
+
| URL 完全無法解析 | `Rejected deep link: not a parseable URL` | 檢查有沒有漏編碼、字串是不是完整的 URL |
|
|
482
|
+
| scheme 不是 `kenaz` | `Rejected deep link: unsupported scheme: <scheme>` | 確認開頭是 `kenaz://` |
|
|
483
|
+
| host 不是 `connect` | `Rejected deep link: not a connect action` | 確認是 `kenaz://connect?...`,不是別的 action |
|
|
484
|
+
| 缺少必填欄位 | `Rejected deep link: missing field: site` / `mcp` / `token` | 檢查第 2 節參數表——`site`/`mcp`/`token` 三個都必填 |
|
|
485
|
+
| `site` 不是合法 URL | `Rejected deep link: site is not a parseable URL` | 確認 `site=` 是一個完整、可解析的 URL |
|
|
486
|
+
| `site` scheme 不允許 | `Rejected deep link: site scheme not allowed: http` | 正式環境用 `https://`;本機測試需要開發模式(第 5 節)且 host 必須是 `localhost`/`127.0.0.1`/`[::1]` 其中之一(bracket 形式的 `[::1]`,其他 IPv6 位址不算) |
|
|
487
|
+
| `mcp` 不是合法 URL | `Rejected deep link: mcp is not a parseable URL` | 確認 `mcp=` 是一個完整、可解析的 URL |
|
|
488
|
+
| `mcp` scheme 不允許 | `Rejected deep link: mcp scheme not allowed: http` | 同上,`mcp=` 適用一模一樣的 https/開發模式規則 |
|
|
489
|
+
| `token` 是空字串 | `Rejected deep link: token is empty` | `token=` 不能是空值 |
|
|
490
|
+
| `token` 超過 512 bytes | `Rejected deep link: token too long: <n> bytes` | 縮短你的 token;大內容走第 3.4 節「大檔案走 URL」 |
|
|
491
|
+
| `token` 含不合法字元 | `Rejected deep link: token has invalid characters` | 字元集只能是 `[A-Za-z0-9._-]` |
|
|
492
|
+
| `v` 不是 Kenaz 支援的版本 | `Rejected deep link: protocol version unsupported: <v>` | 目前只支援 `v=1`(或省略);不要自己編一個版本號 |
|
|
493
|
+
| Agent Hub 被使用者暫停(tray「Pause Agent Hub」) | `Agent Hub is paused` | 這是使用者自己的選擇,你的網站無法繞過;只能等使用者自己恢復 |
|
|
494
|
+
| origin 是 `blocked` 狀態 | `<你的 site origin> is blocked` | 見第 4 節「被封鎖後怎麼辦」——只有使用者自己能解除 |
|
|
495
|
+
| 使用者在審批卡按了「拒絕」 | (不會 toast——只是安靜地不執行;`Decision::Deny`) | 這是正常拒絕流程,不是錯誤;你的網站一樣只能靠逾時偵測 |
|
|
496
|
+
| Launch 本身失敗(審批通過**之後**) | `[<error_code>] <錯誤訊息>`,例如 `[agent_unavailable] requested agent "codex" is not installed (probed \`codex --version\`)` 或 `[not_logged_in] codex is not logged in — run \`codex login\` before this site can generate anything` | 通常是使用者本機環境問題(CLI 沒裝、沒登入),不是你網站端能修的;`agent=` 若指定了不存在的 CLI 會落在這裡,不是落在深連結驗證階段 |
|
|
497
|
+
|
|
498
|
+
**重要**:以上全部只在 Kenaz 本機顯示,你的網站永遠看不到。你能做的只有
|
|
499
|
+
第 4 節講的逾時偵測 + fallback 頁面,以及在你自己的文件裡引導使用者去看
|
|
500
|
+
Kenaz 視窗的 toast。
|
|
501
|
+
|
|
502
|
+
### 常見錯誤
|
|
503
|
+
|
|
504
|
+
- **「按了連接沒反應」**:先確認瀏覽器有沒有跳出「要不要用 Kenaz 開啟這個
|
|
505
|
+
連結」的系統對話框——大部分瀏覽器對未安裝的自訂 scheme 是完全靜默的,
|
|
506
|
+
不會有任何視覺回饋。用第 4 節的逾時偵測,不要假設「沒反應」等於使用者
|
|
507
|
+
看到了什麼。
|
|
508
|
+
- **「dev 模式設了環境變數還是被拒絕 `site scheme not allowed: http`」**
|
|
509
|
+
:Kenaz 沒有完整重啟過(見第 5 節——環境變數只在行程啟動當下讀一次,
|
|
510
|
+
必須先 tray Quit 再重開)。
|
|
511
|
+
- **「MCP 一直收不到 `kenaz_hello`,但審批卡有出現而且我按了允許」**:先
|
|
512
|
+
用第 6 節的 curl 手法確認你的 MCP 端點本身沒問題(bearer 對得上、
|
|
513
|
+
`kenaz_hello` 回應形狀正確)。如果 MCP 端點本身沒問題,通常是 agent
|
|
514
|
+
端啟動失敗(見上面「Launch 本身失敗」那一列的 toast),跟你的網站/MCP
|
|
515
|
+
無關。
|
|
516
|
+
- **「同一個 origin 明明按過永遠允許,這次又跳審批卡」**:檢查是不是換了
|
|
517
|
+
`mcp=` 的 hostname 或整個換了瀏覽器/使用者 profile——四態權限的 key 是
|
|
518
|
+
`site=` 的 origin,不是 `mcp=`,正常情況下換 `mcp=` 不會重置狀態,但如
|
|
519
|
+
果連 `site=` 都變了(例如從 `https://your-site.example` 換成
|
|
520
|
+
`https://www.your-site.example`)就是不同 origin,會重新從
|
|
521
|
+
`unauthorized` 開始。
|
|
522
|
+
|
|
523
|
+
---
|
|
524
|
+
|
|
525
|
+
## 8. 安全
|
|
526
|
+
|
|
527
|
+
- **Token 不可重用**:每次「連接」都簽一個新的一次性 token,不要把同一個
|
|
528
|
+
值發給下一次連接請求——重用會讓你失去「這個 token 對應哪一次連接」的可
|
|
529
|
+
追蹤性,也讓撤銷變得沒有意義。
|
|
530
|
+
- **不要把長期憑證放進深連結**:深連結的 `token=` 只應該是一個短命、限定
|
|
531
|
+
用途的 bearer,绑定這一次連接。不要把使用者的長期 API key、OAuth
|
|
532
|
+
refresh token 之類的東西直接塞進 `token=`——深連結會經過 OS 的 URL
|
|
533
|
+
handler,也可能被瀏覽器歷史記錄或系統日誌留下痕跡。
|
|
534
|
+
- **MCP 只信 bearer,不要信任何其他隱含的信任來源**:Kenaz 不會替你驗證
|
|
535
|
+
`token`;你的 MCP 端點收到的每一個請求都要自己驗證
|
|
536
|
+
`Authorization: Bearer <token>`,不要假設「請求是從 Kenaz 打來的所以一
|
|
537
|
+
定合法」——任何知道你 MCP URL 的人都可以直接對它送 HTTP 請求,繞過整個
|
|
538
|
+
深連結流程。
|
|
539
|
+
|
|
540
|
+
---
|
|
541
|
+
|
|
542
|
+
## 9. 持續對話:`kenaz_wait_for_message`(TASK_1948,FlowPix S2)
|
|
543
|
+
|
|
544
|
+
第 3 節的 `kenaz_hello` 只解決「連上了沒」——如果你的網站想在那之後**繼續
|
|
545
|
+
跟 agent 對話**(例如你自己的聊天 UI 使用者又打了一行字,或畫布又產生了新
|
|
546
|
+
的狀態要交給 agent 處理),過去唯一的做法是自己在 `prompt=` 或某個工具的
|
|
547
|
+
回應裡塞一堆「請持續呼叫我」的 prompt 工程指令,效果不穩定也不可測試。
|
|
548
|
+
v0.4 把這件事收斂成一個**可選的**約定工具,讓「agent 會不會一直聽」這件事
|
|
549
|
+
從你的 prompt 措辭搬進 Kenaz 人格本身。
|
|
550
|
+
|
|
551
|
+
### 這是什麼
|
|
552
|
+
|
|
553
|
+
你的 MCP 若在 `tools/list` 多提供一個無參數工具 `kenaz_wait_for_message`,
|
|
554
|
+
Kenaz 注入的 Ansuz 人格會在呼叫過 `kenaz_hello`、處理完 `prompt` 之後,
|
|
555
|
+
**自動反覆呼叫它**,不需要你在 prompt 裡寫任何「請持續輪詢」之類的指
|
|
556
|
+
令——完全不提供這個工具的網站,行為與現在完全一樣(agent 只呼叫一次
|
|
557
|
+
`kenaz_hello`,之後就照 prompt 跟你其餘的工具互動,不會提起、也不會呼叫
|
|
558
|
+
一個不存在的工具)。
|
|
559
|
+
|
|
560
|
+
**`prompt` 省略時由 Kenaz 保證**(TASK_1956):CLI 的第一個位置參數就是互
|
|
561
|
+
動式會話的第一則使用者訊息,省略/空字串在機制上等於「沒有第一個
|
|
562
|
+
turn」——連 Kenaz 注入的人格本身都不會被 agent 讀到,`kenaz_hello`/
|
|
563
|
+
`kenaz_wait_for_message` 這整套約定自然也就永遠不會發生。Kenaz 因此在
|
|
564
|
+
`prompt` 為空/空白時,自己補一句純 ASCII 的開場訊息(帶你的 `site`
|
|
565
|
+
origin)當作這第一則訊息,讓上面這段流程照樣跑起來。**這代表你完全不需
|
|
566
|
+
要、也不應該**為了讓 agent「動起來」而自己塞一句 `prompt=` 當觸發
|
|
567
|
+
器——那正是這個約定原本要讓你省掉的 prompt 工程。
|
|
568
|
+
|
|
569
|
+
### 回應形狀
|
|
570
|
+
|
|
571
|
+
```json
|
|
572
|
+
{"jsonrpc": "2.0", "id": 2, "method": "tools/call", "params": {"name": "kenaz_wait_for_message", "arguments": {}}}
|
|
573
|
+
```
|
|
574
|
+
|
|
575
|
+
```json
|
|
576
|
+
{"jsonrpc": "2.0", "id": 2, "result": {"content": [{"type": "text", "text": "{\"message\":\"add a red circle to the canvas\",\"message_id\":\"m_1\",\"end\":false}"}], "isError": false}}
|
|
577
|
+
```
|
|
578
|
+
|
|
579
|
+
- `message` 非 `null`:agent 把它當成使用者的下一則指令執行,執行完再呼叫
|
|
580
|
+
一次。
|
|
581
|
+
- `message` 為 `null`:目前沒有新訊息,不是錯誤——agent 直接再呼叫一次
|
|
582
|
+
(這是**預期的常態**,不是逾時或失敗)。
|
|
583
|
+
- `end` 為 `true`(可省略,預設 `false`):agent 停止呼叫,告知使用者對話
|
|
584
|
+
已結束。
|
|
585
|
+
- agent 連續呼叫這個工具失敗(逾時、拋錯、連線失敗)達 **3 次**,也會停
|
|
586
|
+
止並告知使用者——你不需要自己實作「停止」邏輯之外的任何額外信號。
|
|
587
|
+
|
|
588
|
+
完整的權威規格見 `./PROTOCOL.md` 第 2 節第 5 點;`examples/minimal-site/
|
|
589
|
+
server.js` 有一份可跑的參考實作(記憶體信箱:`POST /message` 投遞,
|
|
590
|
+
`kenaz_wait_for_message` 長輪詢拉取,見該範例的 README「持續對話」一節)。
|
|
591
|
+
|
|
592
|
+
### 用長輪詢實作,不是短輪詢——這個成本算在使用者頭上
|
|
593
|
+
|
|
594
|
+
**這是本節唯一的強制條款**:`kenaz_wait_for_message` 的每一次呼叫都是一次
|
|
595
|
+
真實的 agent turn,成本由**呼叫它的 Kenaz 使用者自己的訂閱額度**承擔,不
|
|
596
|
+
是你網站的成本。這代表:
|
|
597
|
+
|
|
598
|
+
- 你的伺服器端實作**必須**用長輪詢——收到請求後,如果暫時沒有新訊息,先
|
|
599
|
+
阻塞著(建議 **20-25 秒**)等新訊息進來,逾時了才回 `message: null`;
|
|
600
|
+
不要立刻回應。
|
|
601
|
+
- **不得**用短輪詢(例如每 1 秒就回一次 `null`,靠 agent 自己在 1 秒後再
|
|
602
|
+
呼叫一次)——這會讓 agent 為了等你的下一則訊息,高頻率地消耗使用者的
|
|
603
|
+
訂閱額度,而這個消耗對你的網站完全免費、對使用者卻是真金白銀。把長輪
|
|
604
|
+
詢窗口拉長,是你對使用者訂閱額度的責任,不是效能優化的選項。
|
|
605
|
+
|
|
606
|
+
### 跟 FlowPix 自己的信箱模式的對照
|
|
607
|
+
|
|
608
|
+
如果你已經有一套「使用者訊息 → 佇列 → worker 拉取處理」的信箱/佇列架
|
|
609
|
+
構(FlowPix 內部原本就是這樣:畫布事件先進一個佇列,worker 再消費),
|
|
610
|
+
`kenaz_wait_for_message` 幾乎是直接把你信箱的**消費端**換成「agent 呼叫
|
|
611
|
+
的 MCP 工具」而已:
|
|
612
|
+
|
|
613
|
+
| FlowPix 原有信箱模式 | `kenaz_wait_for_message` |
|
|
614
|
+
|---|---|
|
|
615
|
+
| worker 對佇列做長輪詢 / blocking pop | agent 對你的 MCP 工具做長輪詢式呼叫 |
|
|
616
|
+
| 佇列裡的一筆訊息 | 回應裡的 `message` 欄位 |
|
|
617
|
+
| 訊息的唯一 ID(供去重/確認) | `message_id` 欄位 |
|
|
618
|
+
| worker 收到「關閉佇列」信號就停止消費 | 回應裡的 `end: true` |
|
|
619
|
+
| 佇列基礎設施掛了,worker 重試幾次後放棄 | 連續呼叫失敗 3 次,agent 放棄並告知使用者 |
|
|
620
|
+
|
|
621
|
+
差異只在於:你的信箱現在多了一個**消費者是 Kenaz 使用者本人的 agent
|
|
622
|
+
turn**,每次消費都有真實成本(見上一節)——這也是為什麼即使你已經有一套
|
|
623
|
+
成熟的信箱架構,也不建議把它原本對內部 worker 用的短輪詢頻率原封不動套
|
|
624
|
+
用到這個工具上。
|
|
625
|
+
|
|
626
|
+
### 這節不涵蓋什麼
|
|
627
|
+
|
|
628
|
+
- **不涵蓋 agent 輸出串流回你的網站**(S3,未分期)——`kenaz_wait_for_
|
|
629
|
+
message` 只是「網站 → agent」單向的訊息投遞,agent 執行過程或結果不會
|
|
630
|
+
透過這個機制串流回你的網站;如果你需要看到執行進度,一樣得靠你自己另
|
|
631
|
+
外設計的工具(例如 agent 主動呼叫的一個「回報進度」工具)。
|
|
632
|
+
- **不涵蓋 agent 主動通知「我結束了」**(S4,未分期)——`end: true` 只是
|
|
633
|
+
你告訴 agent「別再問了」,不是 agent 告訴你「我做完了」;agent 結束一
|
|
634
|
+
段對話(不管是你送了 `end: true`,還是使用者自己叫它停)之後,終端分頁
|
|
635
|
+
何時關閉、agent 進程是否還活著,你的網站依然無從得知(第 3.4 節「沒有
|
|
636
|
+
結束通知」的限制在這裡同樣成立)。
|
|
637
|
+
- **不改變審批卡文案**——目前的審批卡(第 4 節)不會因為你的 MCP 有沒有
|
|
638
|
+
`kenaz_wait_for_message` 而顯示任何額外內容。
|
|
639
|
+
|
|
640
|
+
---
|
|
641
|
+
|
|
642
|
+
## 10. 未安裝流程:偵測 + 導向安裝頁(TASK_1950)
|
|
643
|
+
|
|
644
|
+
第 4 節「請安裝 Kenaz 的偵測與 fallback」講的是**一般性**的逾時
|
|
645
|
+
fallback——60 秒沒收到 `kenaz_hello` 就顯示一段安裝指引文字,而且刻意把
|
|
646
|
+
「沒裝」「使用者拒絕審批」「使用者封鎖了你的網站」三種情況混在一起,因為
|
|
647
|
+
網站真的分辨不出來(第 0 節「沒有回程通道」)。
|
|
648
|
+
|
|
649
|
+
這一節加的是一個**更快、專門針對「可能沒裝」的獨立偵測層**,用途不是取
|
|
650
|
+
代第 4 節的 60 秒一般性 fallback,而是在它之前先做一次快速判斷、**主動導
|
|
651
|
+
覽**到安裝頁(不只是顯示一段文字),並把原始深連結透過 `return=` 帶過
|
|
652
|
+
去,讓使用者裝完 Kenaz 之後不用再按一次「連接」。`@kenaz/connect` 這個
|
|
653
|
+
SDK npm 套件本身**尚未實作**(C 期產出,見 §6「目前沒有一個『假
|
|
654
|
+
Kenaz』npm 套件」段落同樣的分期狀態)——這一節記錄的是它的目標行為,
|
|
655
|
+
`examples/minimal-site/` 用零依賴的原生 JS 先示範同一套邏輯,供現在就要
|
|
656
|
+
接入的網站參考。
|
|
657
|
+
|
|
658
|
+
### 流程圖
|
|
659
|
+
|
|
660
|
+
```
|
|
661
|
+
使用者按「連接」
|
|
662
|
+
│
|
|
663
|
+
▼
|
|
664
|
+
組深連結 kenaz://connect?...(第 2 節)→ 記下 startedAt
|
|
665
|
+
│
|
|
666
|
+
▼
|
|
667
|
+
window.location.href = 深連結(同一分頁導覽,OS 靜默失敗或開啟 Kenaz)
|
|
668
|
+
│
|
|
669
|
+
▼
|
|
670
|
+
開始輪詢自己後端的連線狀態(例如 GET /api/status/:token)──┐
|
|
671
|
+
│ │
|
|
672
|
+
│ 每次輪詢檢查: │ 每 1 秒
|
|
673
|
+
▼ │
|
|
674
|
+
已收到 kenaz_hello?──是──> 顯示「已連線」,結束 │
|
|
675
|
+
│否 │
|
|
676
|
+
▼ │
|
|
677
|
+
now - startedAt >= NOT_INSTALLED_TIMEOUT_MS(預設 8 秒)?──否─┘
|
|
678
|
+
│是
|
|
679
|
+
▼
|
|
680
|
+
導覽到安裝頁:installUrl + "?return=" + encodeURIComponent(原始深連結)
|
|
681
|
+
│
|
|
682
|
+
▼
|
|
683
|
+
(使用者在安裝頁下載、安裝、Kenaz 完成 scheme 註冊)
|
|
684
|
+
│
|
|
685
|
+
▼
|
|
686
|
+
安裝頁重新觸發 return= 裡的那條深連結
|
|
687
|
+
│
|
|
688
|
+
▼
|
|
689
|
+
回到第 2 節的深連結流程(審批卡 → 終端分頁 → kenaz_hello)
|
|
690
|
+
```
|
|
691
|
+
|
|
692
|
+
### 逾時判定:8 秒預設,backend 沒收到 `kenaz_hello` 為準
|
|
693
|
+
|
|
694
|
+
- **預設 8 秒**(`examples/minimal-site` 裡的 `NOT_INSTALLED_TIMEOUT_MS`,
|
|
695
|
+
可調——見下方「minimal-site 怎麼示範」)。比第 4 節的 60 秒短得多,因
|
|
696
|
+
為這裡的目的是快速把「大概率沒裝」的使用者送去安裝頁,不是給審批卡的
|
|
697
|
+
使用者充分反應時間。
|
|
698
|
+
- **判定依據**:網站自己的後端在這個時間窗內有沒有收到 agent 呼叫
|
|
699
|
+
`kenaz_hello`(第 3.2 節)——輪詢自己後端的連線狀態端點,跟第 4 節同一
|
|
700
|
+
個機制,只是時間窗更短。
|
|
701
|
+
- **瀏覽器 blur/visibilitychange 不是主要依據**:導覽到 `kenaz://` 這類
|
|
702
|
+
自訂 scheme 時,部分瀏覽器/OS 組合會短暫觸發分頁失焦(跳出「要不要用
|
|
703
|
+
Kenaz 開啟」的系統對話框,或 OS 真的切到 Kenaz 視窗),可以拿來當一個
|
|
704
|
+
**加速捷徑**(例如「分頁在 1 秒內重新拿到焦點」,八成是系統對話框被使
|
|
705
|
+
用者選了「取消」,可以提早判定沒裝)——但這個訊號在不同瀏覽器、不同
|
|
706
|
+
OS 上是否觸發、觸發時機都不一致,**不能當成唯一或主要判定依據**,`kenaz_hello`
|
|
707
|
+
逾時仍然是唯一可靠的來源。
|
|
708
|
+
- **已知的取捨(未解決,如實記錄)**:8 秒對於「已經裝了 Kenaz,但使用
|
|
709
|
+
者還沒看到/還沒按審批卡」的使用者是會誤判的——審批卡跳出、使用者切
|
|
710
|
+
去讀別的視窗、8 秒內沒按「允許」,一樣會被導去安裝頁。這是這個設計故
|
|
711
|
+
意接受的假陽性,換取「真的沒裝」的使用者不用等到 60 秒才知道要去哪
|
|
712
|
+
裡。如果你的網站場景使用者反應通常較慢,把
|
|
713
|
+
`NOT_INSTALLED_TIMEOUT_MS`/`KENAZ_INSTALL_TIMEOUT_MS` 調大一些即可——這
|
|
714
|
+
就是它必須是可調參數、不能寫死的原因。
|
|
715
|
+
|
|
716
|
+
### 安裝頁 URL 從哪裡來——這是一個可設定參數,不是臆造的路由
|
|
717
|
+
|
|
718
|
+
**目前沒有已知的 `app.kenazai.com` 安裝頁路由**,本檔不臆造一個。查證
|
|
719
|
+
`dashboard/scripts/publish-release.js` 得到的實際事實:
|
|
720
|
+
|
|
721
|
+
- Kenaz 的正式安裝檔上傳到 Supabase Storage 的 `kenaz-release` bucket
|
|
722
|
+
(`publish-release.js` 第 19 行 `const BUCKET = 'kenaz-release'`)。
|
|
723
|
+
- 下載 URL 形狀:
|
|
724
|
+
`{SUPABASE_URL}/storage/v1/object/public/kenaz-release/{version}/{encodeURIComponent(nsisZip檔名)}`
|
|
725
|
+
(第 113-115 行)。
|
|
726
|
+
- `latest.json` 上傳到同一個 bucket 的根目錄(`{SUPABASE_URL}/storage/v1/
|
|
727
|
+
object/public/kenaz-release/latest.json`),內容形狀是
|
|
728
|
+
`{version, notes, pub_date, platforms: {"windows-x86_64": {url,
|
|
729
|
+
signature}}}`(第 117-127 行)——這是 Tauri updater 消費的格式,不是給
|
|
730
|
+
瀏覽器直接開的人類可讀安裝頁。
|
|
731
|
+
- `SUPABASE_URL` 本身**不在本 repo 裡**(`publish-release.js` 從
|
|
732
|
+
gitignored 的 `.env.local` 讀取,第 21-35 行)——因此本檔**無法引用**
|
|
733
|
+
正式環境的實際 bucket URL,只能引用上面這個結構事實。
|
|
734
|
+
|
|
735
|
+
**結論**:在 Kenaz 提供一個正式的人類可讀安裝頁之前,`installUrl` 是
|
|
736
|
+
SDK/範例的一個**必填設定參數**,不是硬編碼常數:
|
|
737
|
+
|
|
738
|
+
- `examples/minimal-site/server.js` 讀取環境變數 `KENAZ_INSTALL_URL`
|
|
739
|
+
(沒設時退回一個明顯是佔位符的預設值
|
|
740
|
+
`https://kenaz-install.example/placeholder`,`.example` TLD 保證不會意
|
|
741
|
+
外解析成真的網域),伺服器端把它注入 `public/index.html` 連接按鈕的
|
|
742
|
+
`data-install-url` 屬性。
|
|
743
|
+
- 在 Kenaz 提供正式安裝頁之前,網站可以自行把這個參數指向:**(a)** 自
|
|
744
|
+
家的下載頁(自己寫「請安裝 Kenaz」+ 下載連結 + 裝完後手動再按一次「連
|
|
745
|
+
接」),或 **(b)** 一個會去讀 `latest.json`、解析出
|
|
746
|
+
`platforms['windows-x86_64'].url` 再自動觸發下載的輕量頁面。無論哪
|
|
747
|
+
種,`return=` 參數(下一節)都應該由網站自己的頁面在裝完之後負責重新
|
|
748
|
+
觸發,而不是假設 Kenaz 的安裝程式本身會處理 `return=`——目前的 NSIS 安
|
|
749
|
+
裝程式沒有讀取/處理任何啟動參數的已知邏輯(未在本任務驗證範圍內找到相
|
|
750
|
+
反證據,也未逐行查證安裝程式腳本;如果之後要做到「安裝完自動觸發
|
|
751
|
+
`return=`」,需要另外驗證/修改安裝程式,這不在本任務 scope 內)。
|
|
752
|
+
|
|
753
|
+
### `return=`:帶原始深連結,以及一個目前沒有協定層數字保證的 TTL 風險
|
|
754
|
+
|
|
755
|
+
`return=` 的值是**原始的 `kenaz://connect?...` 深連結**,整個字串經
|
|
756
|
+
`URLSearchParams.set('return', deepLink)` 處理(等同
|
|
757
|
+
`encodeURIComponent`,見 `examples/minimal-site/public/install-redirect.js`
|
|
758
|
+
的 `buildInstallRedirectUrl()`)。
|
|
759
|
+
|
|
760
|
+
**查證 TTL 數字的結果(更正一個容易望文生義的假設)**:`PROTOCOL.md` §2
|
|
761
|
+
點 1 與本檔 §3.1 都只寫「TTL:建議短——這個 token 只需要活到這次終端分
|
|
762
|
+
頁的 session 期間」,**沒有任何具體秒數**——Kenaz 完全不驗證、不落盤這個
|
|
763
|
+
token(第 0 節「沒有回程通道」的同一個事實:Kenaz 只是原樣轉交),TTL
|
|
764
|
+
純粹是網站自己的決定,協定層沒有規定任何數字。
|
|
765
|
+
|
|
766
|
+
這對「未安裝流程」有一個實際後果:使用者從按下「連接」到裝完 Kenaz、
|
|
767
|
+
`return=` 裡的深連結被重新觸發,中間可能經過**任意長**的時間(下載、安
|
|
768
|
+
裝、可能還要重開機)——遠超過「一個終端分頁 session 的存續期間」這個一
|
|
769
|
+
般性 TTL 假設所預期的量級。如果網站的 token 依照 §3.1 的一般建議設得很
|
|
770
|
+
短,`return=` 重新觸發時很可能已經過期,導致裝完 Kenaz 後审批卡看起來
|
|
771
|
+
正常跳出、agent 也啟動了,但呼叫 `kenaz_hello` 時網站的 bearer 驗證卻
|
|
772
|
+
401——使用者會卡在「裝都裝了怎麼還是連不上」。
|
|
773
|
+
|
|
774
|
+
**建議(本檔的建議,不是協定規定)**:網站在替「未安裝流程」產生 token
|
|
775
|
+
時,要嘛(a)給這個特定用途的 token 一個明顯更寬容的 TTL(例如以「分
|
|
776
|
+
鐘」而非「這一個 session」為單位),要嘛(b)更穩妥的做法——`return=`
|
|
777
|
+
不要直接指向裸的 `kenaz://` 深連結,而是指回網站自己的一個頁面
|
|
778
|
+
(例如 `https://your-site.example/connect?resume=1`),讓那個頁面重新
|
|
779
|
+
執行第 1 節「步驟一:簽 token」+「步驟二:組深連結」重新簽發一個全新
|
|
780
|
+
token,再導覽到新組出的深連結——這樣就完全不用假設任何 TTL 撐得過使用
|
|
781
|
+
者離開網站的這段時間。`examples/minimal-site` 目前示範的是最直接的
|
|
782
|
+
(a) 類做法的簡化版(沿用原深連結字面值),選 (b) 的網站需要自己加一個
|
|
783
|
+
resume 路由,這不在本任務 scope 內、範例沒有實作。
|
|
784
|
+
|
|
785
|
+
### minimal-site 怎麼示範
|
|
786
|
+
|
|
787
|
+
`examples/minimal-site/public/install-redirect.js` 是這段邏輯的**純函
|
|
788
|
+
數**實作(刻意抽成獨立檔案而不是塞進 `index.html` 的 inline script,因
|
|
789
|
+
為同一份邏輯要同時被瀏覽器跟 `node --test` 匯入,見該檔案的檔頭註解):
|
|
790
|
+
|
|
791
|
+
- `hasInstallTimedOut({ startedAt, timeoutMs, now })`:逾時判定,`now`
|
|
792
|
+
可外部注入方便測試。
|
|
793
|
+
- `buildInstallRedirectUrl(installUrl, deepLink)`:組出帶 `return=` 的安
|
|
794
|
+
裝頁 URL。
|
|
795
|
+
- `resolveInstallRedirect({ connected, startedAt, timeoutMs, installUrl,
|
|
796
|
+
deepLink, now })`:單一決策入口——`connected` 為真或還沒逾時都回
|
|
797
|
+
`null`(`connected` 的檢查優先於逾時檢查,逾時後才收到的「已連線」回
|
|
798
|
+
應不應該還被導去安裝頁),逾時且未連線才回傳導向 URL。
|
|
799
|
+
|
|
800
|
+
`public/index.html` 的連接按鈕透過 `data-install-url` /
|
|
801
|
+
`data-not-installed-timeout-ms` 兩個屬性讀取設定(`server.js` 的
|
|
802
|
+
`renderIndexHtml()` 從 `KENAZ_INSTALL_URL` / `KENAZ_INSTALL_TIMEOUT_MS`
|
|
803
|
+
環境變數注入,見該檔案),輪詢迴圈每次都呼叫
|
|
804
|
+
`resolveInstallRedirect()`,拿到非 `null` 就直接
|
|
805
|
+
`window.location.href = redirectUrl` 導覽過去。
|
|
806
|
+
|
|
807
|
+
`test/smoke.test.mjs` 用 200ms 的短逾時(而不是真的 8 秒)驗證這條路徑
|
|
808
|
+
(TASK_1950 AC1):
|
|
809
|
+
|
|
810
|
+
- 逾時前呼叫 `resolveInstallRedirect` 必須回 `null`。
|
|
811
|
+
- 逾時後(且 `connected: false`)必須回傳一個 URL,其 `return=` 參數解
|
|
812
|
+
碼後**等於原始深連結字串**。
|
|
813
|
+
- 對照組:`connected: true` 時,就算已經遠遠超過逾時,也絕不能回傳導向
|
|
814
|
+
URL(`feedback_probe_needs_control_group` 這條記憶——沒有對照組的通過
|
|
815
|
+
證明不了邏輯是對的,只證明程式碼跑起來沒拋錯)。
|
|
816
|
+
- 額外驗證 `GET /` 真的把設定值注入進 HTML、`GET /install-redirect.js`
|
|
817
|
+
把同一份模組當真的靜態檔案提供(不是另外複製一份給瀏覽器用)。
|
|
818
|
+
|
|
819
|
+
跑法:
|
|
820
|
+
|
|
821
|
+
```bash
|
|
822
|
+
node --test dev/specs/agent-connect/examples/minimal-site/test/smoke.test.mjs
|
|
823
|
+
```
|
|
824
|
+
|
|
825
|
+
### 冷啟動:Windows 上的深連結能不能被剛啟動的 Kenaz 接住?
|
|
826
|
+
|
|
827
|
+
這是本節最需要如實記錄的部分——**已修復(TASK_1953):Kenaz 啟動時讀取
|
|
828
|
+
`get_current` 補處理;待實機冷啟動驗證**,不是「已驗證可行」:
|
|
829
|
+
|
|
830
|
+
1. `tauri-plugin-deep-link` 2.4.10(`Cargo.lock` 釘的版本)在**它自己的
|
|
831
|
+
plugin `.setup()` 階段**(`init_deep_link()`,套件原始碼
|
|
832
|
+
`src/lib.rs` 第 68-79 行)就會呼叫
|
|
833
|
+
`handle_cli_arguments(std::env::args())`(第 196-220 行):Windows/
|
|
834
|
+
Linux 上,如果啟動參數只有一個且能解析成一個符合設定 scheme 的 URL,
|
|
835
|
+
會把它存進 `self.current`(供 `get_current()` 讀取)**並且**立刻
|
|
836
|
+
`self.app.emit("deep-link://new-url", vec![url])`。
|
|
837
|
+
2. `on_open_url()`(同檔案第 514-524 行)的實作就是
|
|
838
|
+
`app.listen("deep-link://new-url", ...)`——一個標準的 Tauri event
|
|
839
|
+
listener,**不會**補送監聽之前就發生過的事件。
|
|
840
|
+
3. Tauri 2.10.2 的 `Builder::build()`(`tauri` crate `src/app.rs` 第
|
|
841
|
+
2094 行起)會呼叫 `app.manager.initialize_plugins(handle)`(第 2286
|
|
842
|
+
行附近),這一步會跑過每個 plugin 自己的 `.setup()`——包含上面第 1 點
|
|
843
|
+
的 `handle_cli_arguments` 呼叫與事件 emit。這整個 `build()` 呼叫,發
|
|
844
|
+
生在 KenazAI 自己那個 app 層級的 `.setup(|app| {...})` 閉包(`lib.rs`
|
|
845
|
+
第 1331 行起)**執行之前**——後者是被 `src/app.rs` 裡另一個自由函式
|
|
846
|
+
`setup<R>()`(第 2371 行 `app.ran_setup = true`)呼叫的,而那個自由函
|
|
847
|
+
式是在 `App::run()` 的事件迴圈裡才觸發,時間點晚於 `build()`。
|
|
848
|
+
4. KenazAI 自己的
|
|
849
|
+
`dashboard/src-tauri/src/features/agent_hub/mod.rs`
|
|
850
|
+
`install_deep_link_handler()`(第 111-139 行)是在上述第 3 點「較晚」
|
|
851
|
+
的那個 `.setup()` 閉包裡才呼叫(`lib.rs` 第 1400 行),而且**只**呼叫
|
|
852
|
+
`on_open_url()`——通篇 `dashboard/src-tauri/src/` 搜尋 `get_current()`
|
|
853
|
+
在 deep-link 這個 extension 上的呼叫,結果是**零筆**。
|
|
854
|
+
|
|
855
|
+
四點合起來:**在 Windows 上,一個真正的「Kenaz 完全沒在跑,直接靠雙擊
|
|
856
|
+
`kenaz://connect?...` 冷啟動」的情境下,`handle_cli_arguments` 在 plugin
|
|
857
|
+
setup 階段 emit 的那個事件,此時還沒有任何 listener 訂閱它(`on_open_
|
|
858
|
+
url` 要等到後面的 app-level `.setup()` 才註冊)——這個事件因此被靜靜地
|
|
859
|
+
丟掉。這是 TASK_1950 查證當下(沒有任何地方呼叫 `get_current()`)的實際
|
|
860
|
+
狀態,會導致不出審批卡、不出 toast、什麼都不會發生——跟「使用者根本沒
|
|
861
|
+
看到任何反應」的現象一模一樣。
|
|
862
|
+
|
|
863
|
+
**已修復(TASK_1953)**:`install_deep_link_handler()`
|
|
864
|
+
(`dashboard/src-tauri/src/features/agent_hub/mod.rs`)在註冊完
|
|
865
|
+
`on_open_url` 之後,補一次 `app.deep_link().get_current()`——若冷啟動時
|
|
866
|
+
plugin setup 階段已經捕獲但尚未被任何 listener 處理的 URL 存在,會逐一
|
|
867
|
+
餵進與 `on_open_url` 完全相同的 `handle_incoming_url()` 路徑(兩者共用
|
|
868
|
+
同一個 `spawn_handle_url` helper,不重複解析邏輯);`get_current()` 回
|
|
869
|
+
`Err`/`None` 時只 `tracing::debug!` 靜默記錄,不影響啟動。已在執行時的
|
|
870
|
+
第二次啟動(single-instance 轉發)走的是另一條路徑(見下段),不會與
|
|
871
|
+
`get_current()` 重複處理同一個 URL。
|
|
872
|
+
|
|
873
|
+
這個修復連同上面 1-4 點的查證,完全基於閱讀 `tauri-plugin-deep-link`
|
|
874
|
+
2.4.10 與 `tauri` 2.10.2 的原始碼(本機 cargo registry cache)加上
|
|
875
|
+
KenazAI 自己的原始碼——**尚未做過真的冷啟動實機測試**(TASK_1953 同樣
|
|
876
|
+
在 worktree 中完成,worktree 規則禁止對正在跑的 live tauri dev 動任何
|
|
877
|
+
東西,也不能啟動一個新的 Kenaz 行程)。單實例轉發(已在執行時的第二次
|
|
878
|
+
啟動)走 `tauri-plugin-single-instance` 的 `deep-link` feature,
|
|
879
|
+
TASK_1928 已實測正常,不受本次修復影響、也不在待驗證範圍內。
|
|
880
|
+
|
|
881
|
+
**實機冷啟動驗證步驟**(待主線程或有桌面環境的人執行):
|
|
882
|
+
|
|
883
|
+
1. 完整關閉所有 Kenaz 行程(含 tray 常駐;dev 模式下也要停掉 `tauri
|
|
884
|
+
dev`,因為 dev 模式的 `register_all()` 會把 `kenaz://` 指到
|
|
885
|
+
`target/debug` 的 exe,需要重新啟動一次乾淨的行程來吃到冷啟動路徑)。
|
|
886
|
+
2. 在瀏覽器網址列直接開啟一個合法的 `kenaz://connect?...` 深連結
|
|
887
|
+
(帶上 `origin`/`mcp`/`token`/`prompt` 等必要參數,見本檔第 2/3 節)。
|
|
888
|
+
3. 預期:OS 拉起 Kenaz 主行程後,視窗前景顯示審批卡(而不是靜默無反應、
|
|
889
|
+
也不是網站端 `kenaz_hello` 逾時後被 TASK_1950 的未安裝流程誤判成
|
|
890
|
+
「未安裝」)。
|
|
891
|
+
4. 按下「允許」後應正常走完 §0 心智模型的第 3-5 步(開終端分頁、agent
|
|
892
|
+
掛上網站的 MCP、網站收到 `kenaz_hello`)。
|
|
893
|
+
|
|
894
|
+
**AC3 判定**:程式碼修復已完成並可 code review 驗證(見上),但「實機冷
|
|
895
|
+
啟動時 OS 真的會如描述般拉起行程並吃到這條路徑」仍是**待驗證**——這裡
|
|
896
|
+
誠實記錄為「已修復,待實機驗證」,不是「已驗證沒問題」。
|
|
897
|
+
|
|
898
|
+
---
|
|
899
|
+
|
|
900
|
+
## 參考
|
|
901
|
+
|
|
902
|
+
- `./PROTOCOL.md`(v0.4,本檔每個行為的權威來源)
|
|
903
|
+
- `./examples/minimal-site/`(可跑範例 + `node --test` 煙霧測試,含
|
|
904
|
+
`kenaz_wait_for_message` 信箱參考實作、§10「未安裝流程」的
|
|
905
|
+
`install-redirect.js` 參考實作)
|
|
906
|
+
- `dev/KENAZ_AGENT_CONNECT_v1_SPEC.md`(規格總覽,含分期與已決定事項)
|
|
907
|
+
- Kenaz 側實作(僅供對照,你不需要讀這些就能接入):
|
|
908
|
+
`dashboard/src-tauri/src/features/agent_hub/{deep_link,mod,launch,permissions,approval}.rs`
|
|
909
|
+
- 假 MCP(Kenaz 自己的 conformance 用,不是你要跑的東西):
|
|
910
|
+
`dashboard/scripts/agent-hub-fake-mcp.mjs`
|
|
911
|
+
- 安裝檔發布腳本(§10「安裝頁 URL 從哪裡來」的事實來源):
|
|
912
|
+
`dashboard/scripts/publish-release.js`
|
|
913
|
+
- deep-link 冷啟動查證(§10 最後一節)的原始碼來源:
|
|
914
|
+
`tauri-plugin-deep-link` 2.4.10(cargo registry cache)、`tauri` 2.10.2
|
|
915
|
+
`src/app.rs`
|