@cbartens/bkit 5.5.1 → 6.1.0
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/.bkit/README.md +167 -154
- package/.bkit/core/.agents/context.md +95 -29
- package/.bkit/core/.agents/docs.md +18 -15
- package/.bkit/core/.agents/plugins/bkit/hooks/bkit-gateguard.cjs +1 -0
- package/.bkit/core/.agents/plugins/bkit/hooks/bkit-guardrails.cjs +6 -5
- package/.bkit/core/.agents/plugins/bkit/hooks/bkit-readguard.cjs +1 -0
- package/.bkit/core/.agents/plugins/bkit/hooks/bkit-rules.cjs +65 -0
- package/.bkit/core/.agents/rules.md +112 -102
- package/.bkit/core/.agents/skills.md +63 -15
- package/.bkit/core/.bkit/README.md +167 -154
- package/.bkit/core/.bkit/authoring.md +747 -0
- package/.bkit/core/.bkit/cli.md +407 -0
- package/.bkit/core/.bkit/cloud.md +283 -195
- package/.bkit/core/.bkit/comparison.md +45 -43
- package/.bkit/core/.bkit/install.md +166 -95
- package/.bkit/core/.bkit/jsonl.md +339 -196
- package/.bkit/core/.bkit/pipeline.md +142 -145
- package/.bkit/core/.bkit/prompts.md +295 -264
- package/.bkit/core/.bkit/providers.md +460 -277
- package/.bkit/core/.bkit/router.md +69 -0
- package/.bkit/core/.bkit/templates.md +530 -256
- package/.bkit/core/.bkit/trackers.md +116 -65
- package/.bkit/core/.bkit/workflows.md +742 -618
- package/.bkit/core/.claude/hooks/bkit-gateguard.cjs +2 -2
- package/.bkit/core/.claude/hooks/bkit-guardrails.cjs +3 -3
- package/.bkit/core/.claude/hooks/bkit-preflight.sh +45 -14
- package/.bkit/core/.codex/hooks/bkit-guardrails.cjs +3 -3
- package/.bkit/core/.codex/hooks/bkit-rules.cjs +57 -0
- package/.bkit/core/.cursor/hooks/bkit-guardrails.cjs +7 -7
- package/.bkit/core/.cursor/hooks/bkit-preflight.sh +47 -12
- package/.bkit/core/.opencode/plugin/bkit-rules.js +89 -0
- package/.bkit/core/AGENTS.md +6 -80
- package/.bkit/core/docs/templates/bkit-audit-accessibility.md +128 -130
- package/.bkit/core/docs/templates/bkit-audit-adoption.md +104 -94
- package/.bkit/core/docs/templates/bkit-audit-agents.md +96 -98
- package/.bkit/core/docs/templates/bkit-audit-aio-seo.md +100 -103
- package/.bkit/core/docs/templates/bkit-audit-analytics.md +141 -107
- package/.bkit/core/docs/templates/bkit-audit-annotations.md +99 -97
- package/.bkit/core/docs/templates/bkit-audit-app-performance.md +122 -121
- package/.bkit/core/docs/templates/bkit-audit-app-store.md +158 -101
- package/.bkit/core/docs/templates/bkit-audit-architecture.md +114 -96
- package/.bkit/core/docs/templates/bkit-audit-ci-performance.md +76 -97
- package/.bkit/core/docs/templates/bkit-audit-config.md +123 -101
- package/.bkit/core/docs/templates/bkit-audit-cost.md +86 -99
- package/.bkit/core/docs/templates/bkit-audit-coverage.md +76 -93
- package/.bkit/core/docs/templates/bkit-audit-database.md +102 -95
- package/.bkit/core/docs/templates/bkit-audit-dead-code.md +165 -108
- package/.bkit/core/docs/templates/bkit-audit-deferrals.md +115 -109
- package/.bkit/core/docs/templates/bkit-audit-dependencies.md +265 -113
- package/.bkit/core/docs/templates/bkit-audit-documentation.md +119 -118
- package/.bkit/core/docs/templates/bkit-audit-duplication.md +140 -111
- package/.bkit/core/docs/templates/bkit-audit-failures.md +111 -106
- package/.bkit/core/docs/templates/bkit-audit-flakes.md +100 -99
- package/.bkit/core/docs/templates/bkit-audit-gates.md +107 -103
- package/.bkit/core/docs/templates/bkit-audit-issues.md +105 -105
- package/.bkit/core/docs/templates/bkit-audit-llm.md +134 -85
- package/.bkit/core/docs/templates/bkit-audit-logs.md +124 -125
- package/.bkit/core/docs/templates/bkit-audit-models.md +99 -78
- package/.bkit/core/docs/templates/bkit-audit-privacy.md +56 -76
- package/.bkit/core/docs/templates/bkit-audit-prompts.md +63 -78
- package/.bkit/core/docs/templates/bkit-audit-red-team.md +125 -93
- package/.bkit/core/docs/templates/bkit-audit-reliability.md +97 -96
- package/.bkit/core/docs/templates/bkit-audit-repo.md +170 -98
- package/.bkit/core/docs/templates/bkit-audit-security.md +207 -121
- package/.bkit/core/docs/templates/bkit-audit-slop.md +78 -95
- package/.bkit/core/docs/templates/bkit-audit-testing.md +163 -139
- package/.bkit/core/docs/templates/bkit-audit-usability.md +114 -96
- package/.bkit/core/docs/templates/bkit-plan.md +158 -281
- package/.bkit/core/engine/checks/after/auth.ts +40 -37
- package/.bkit/core/engine/checks/after/ci-yml.ts +104 -31
- package/.bkit/core/engine/checks/after/github.ts +95 -17
- package/.bkit/core/engine/checks/after/index.ts +21 -8
- package/.bkit/core/engine/checks/after/suppression.ts +8 -5
- package/.bkit/core/engine/checks/after/tools.ts +125 -123
- package/.bkit/core/engine/checks/before/context-fetch.ts +227 -33
- package/.bkit/core/engine/checks/before/env.ts +3 -3
- package/.bkit/core/engine/checks/before/files.ts +5 -4
- package/.bkit/core/engine/checks/before/index.ts +25 -1
- package/.bkit/core/engine/checks/before/repo.ts +208 -2
- package/.bkit/core/engine/checks/status-holes.ts +916 -123
- package/.bkit/core/engine/checks/tips/catalogue.ts +82 -20
- package/.bkit/core/engine/checks/tips/detect.ts +66 -0
- package/.bkit/core/engine/checks/tips/index.ts +52 -11
- package/.bkit/core/engine/cli.ts +251 -130
- package/.bkit/core/engine/commands/analysis.ts +282 -199
- package/.bkit/core/engine/commands/ci/ci.ts +363 -29
- package/.bkit/core/engine/commands/ci/pins.ts +172 -0
- package/.bkit/core/engine/commands/ci/push-gate.ts +223 -0
- package/.bkit/core/engine/commands/comply.ts +77 -103
- package/.bkit/core/engine/commands/install/doctor.ts +2 -2
- package/.bkit/core/engine/commands/install/install.ts +57 -18
- package/.bkit/core/engine/commands/install/uninstall.ts +86 -55
- package/.bkit/core/engine/commands/install/upgrade.ts +32 -15
- package/.bkit/core/engine/commands/land.ts +139 -18
- package/.bkit/core/engine/commands/lib.ts +381 -22
- package/.bkit/core/engine/commands/registry.ts +10 -2
- package/.bkit/core/engine/commands/setup/setup.ts +8 -7
- package/.bkit/core/engine/commands/stats.ts +260 -0
- package/.bkit/core/engine/commands/tips.ts +10 -5
- package/.bkit/core/engine/commands/workflow/branches.ts +2215 -176
- package/.bkit/core/engine/commands/workflow/challenge.ts +748 -0
- package/.bkit/core/engine/commands/workflow/completeness.ts +109 -0
- package/.bkit/core/engine/commands/workflow/context.ts +1249 -47
- package/.bkit/core/engine/commands/workflow/counters.ts +483 -95
- package/.bkit/core/engine/commands/workflow/decisions.ts +585 -0
- package/.bkit/core/engine/commands/workflow/disk.ts +238 -56
- package/.bkit/core/engine/commands/workflow/fids.ts +811 -33
- package/.bkit/core/engine/commands/workflow/issues.ts +413 -64
- package/.bkit/core/engine/commands/workflow/ledgers.ts +1333 -36
- package/.bkit/core/engine/commands/workflow/locks.ts +28 -2
- package/.bkit/core/engine/commands/workflow/plan-blocks.ts +691 -0
- package/.bkit/core/engine/commands/workflow/plan-versions.ts +21 -0
- package/.bkit/core/engine/commands/workflow/plans.ts +257 -0
- package/.bkit/core/engine/commands/workflow/prs.ts +178 -33
- package/.bkit/core/engine/commands/workflow/repo.ts +66 -8
- package/.bkit/core/engine/commands/workflow/roadmap.ts +72 -22
- package/.bkit/core/engine/commands/workflow/run-facts.ts +492 -0
- package/.bkit/core/engine/commands/workflow/runs.ts +827 -79
- package/.bkit/core/engine/commands/workflow/runtime.ts +25 -1
- package/.bkit/core/engine/commands/workflow/skills.ts +1 -0
- package/.bkit/core/engine/commands/workflow/status.ts +329 -33
- package/.bkit/core/engine/commands/workflow/step-table.ts +381 -0
- package/.bkit/core/engine/commands/workflow/steps.ts +841 -105
- package/.bkit/core/engine/commands/workflow/summary-findings.ts +177 -0
- package/.bkit/core/engine/commands/workflow/summary.ts +544 -0
- package/.bkit/core/engine/commands/workflow/tasks.ts +409 -0
- package/.bkit/core/engine/commands/workflow/test-blocks.ts +254 -0
- package/.bkit/core/engine/commands/workflow/tests-changed.ts +201 -0
- package/.bkit/core/engine/commands/workflow/tests.ts +2124 -330
- package/.bkit/core/engine/commands/workflow/tracker.ts +1195 -0
- package/.bkit/core/engine/commands/workflow/workflows.ts +1197 -235
- package/.bkit/core/engine/commands/workflow/worktrees.ts +543 -24
- package/.bkit/core/engine/install/cloud-env.ts +251 -121
- package/.bkit/core/engine/install/context-ci-workflow.ts +102 -0
- package/.bkit/core/engine/install/manifest.ts +67 -17
- package/.bkit/core/engine/install/modules/antigravity.ts +3 -2
- package/.bkit/core/engine/install/modules/cloud.ts +24 -16
- package/.bkit/core/engine/install/modules/codex.ts +15 -2
- package/.bkit/core/engine/install/modules/core.ts +1196 -45
- package/.bkit/core/engine/install/modules/cursor.ts +6 -6
- package/.bkit/core/engine/install/modules/opencode.ts +18 -6
- package/.bkit/core/engine/install/template-notes.ts +130 -0
- package/.bkit/core/engine/install/version-file.ts +9 -4
- package/.bkit/core/engine/install/write.ts +5 -0
- package/.bkit/core/engine/ledgers/compact.ts +124 -0
- package/.bkit/core/engine/ledgers/journal.ts +291 -0
- package/.bkit/core/engine/ledgers/migrate.ts +92 -0
- package/.bkit/core/engine/ledgers/status.ts +249 -21
- package/.bkit/core/engine/ledgers/telemetry.ts +1397 -253
- package/.bkit/core/engine/render/hosts.ts +118 -14
- package/.bkit/core/engine/render/models.ts +12 -0
- package/.bkit/core/engine/shared/bkit-latest.ts +26 -4
- package/.bkit/core/engine/shared/branches.ts +4 -1
- package/.bkit/core/engine/shared/dep-updater.ts +60 -0
- package/.bkit/core/engine/shared/egress.ts +8 -1
- package/.bkit/core/engine/shared/json.ts +25 -7
- package/.bkit/core/engine/shared/package-manager.ts +9 -0
- package/.bkit/core/engine/shared/settings.ts +243 -31
- package/.bkit/core/engine/shared/text.ts +361 -0
- package/.bkit/core/engine/shared/timeouts.ts +2 -0
- package/.bkit/core/engine/tools/cli-doc.ts +111 -0
- package/.bkit/core/engine/tools/grammar-density.ts +215 -0
- package/.bkit/core/engine/tools/instructions-sources.ts +832 -0
- package/.bkit/core/engine/tools/instructions.ts +1802 -0
- package/.bkit/core/hooks/antigravity-hook.js +25 -10
- package/.bkit/core/hooks/gateguard-engine.js +54 -29
- package/.bkit/core/hooks/guardrails-shell.js +4 -5
- package/.bkit/core/hooks/guardrails-tiers.js +10 -4
- package/.bkit/core/hooks/preflight-linear.sh +9 -9
- package/.bkit/core/hooks/preflight-rules.sh +8 -0
- package/.bkit/core/hooks/readguard-engine.js +29 -10
- package/.bkit/core/hooks/session-rules-advisory.js +61 -0
- package/.bkit/core/prompts/bkit-audit-generate.md +202 -94
- package/.bkit/core/prompts/bkit-build-challenge-1.md +76 -34
- package/.bkit/core/prompts/bkit-build-challenge-2.md +85 -38
- package/.bkit/core/prompts/bkit-build-install-skills.md +50 -28
- package/.bkit/core/prompts/bkit-build-plan-convert.md +85 -0
- package/.bkit/core/prompts/bkit-build-plan-current.md +59 -36
- package/.bkit/core/prompts/bkit-build-plan-gaps-remediate.md +101 -36
- package/.bkit/core/prompts/bkit-build-plan-tasks.md +149 -0
- package/.bkit/core/prompts/bkit-build-plan-tests.md +120 -0
- package/.bkit/core/prompts/bkit-build-plan-verify.md +104 -43
- package/.bkit/core/prompts/bkit-debug-abort-handover.md +47 -32
- package/.bkit/core/prompts/bkit-debug-failure-diagnose.md +109 -57
- package/.bkit/core/prompts/bkit-debug-failure-remediate.md +65 -39
- package/.bkit/core/prompts/bkit-debug-failure-reproduce.md +53 -31
- package/.bkit/core/prompts/bkit-debug-failure-soak.md +65 -34
- package/.bkit/core/prompts/bkit-design-brainstorm-ideate-1.md +46 -31
- package/.bkit/core/prompts/bkit-design-brainstorm-ideate-2.md +45 -31
- package/.bkit/core/prompts/bkit-design-brainstorm-ideate-3.md +43 -30
- package/.bkit/core/prompts/bkit-design-generate.md +80 -46
- package/.bkit/core/prompts/bkit-design-organise.md +83 -32
- package/.bkit/core/prompts/bkit-design-review-1.md +70 -50
- package/.bkit/core/prompts/bkit-design-review-2.md +65 -46
- package/.bkit/core/prompts/bkit-design-spike.md +64 -40
- package/.bkit/core/prompts/bkit-design-update.md +53 -28
- package/.bkit/core/prompts/bkit-plan-generate.md +101 -29
- package/.bkit/core/prompts/bkit-plan-review-1.md +78 -41
- package/.bkit/core/prompts/bkit-plan-review-2.md +78 -35
- package/.bkit/core/prompts/bkit-plan-scope-check.md +116 -0
- package/.bkit/core/prompts/bkit-plan-update.md +86 -31
- package/.bkit/core/prompts/bkit-ship-conflict-resolve.md +94 -50
- package/.bkit/core/prompts/bkit-ship-deferred-sweep.md +91 -41
- package/.bkit/core/prompts/bkit-ship-update-docs.md +45 -24
- package/.bkit/core/prompts/bkit-utility-compress-markdown.md +65 -49
- package/.bkit/core/prompts/bkit-utility-update-decisions.md +87 -42
- package/.bkit/core/prompts/bkit-utility-update-deferrals.md +81 -37
- package/.bkit/core/prompts/bkit-utility-update-documentation.md +43 -25
- package/.bkit/core/prompts/bkit-utility-update-failures.md +84 -37
- package/.bkit/core/prompts/bkit-utility-update-knowledge.md +88 -42
- package/.bkit/core/prompts/bkit-utility-update-roadmap.md +92 -39
- package/.bkit/core/prompts/bkit-utility-update-tracker.md +159 -60
- package/.bkit/core/workflows/bkit-auto.md +89 -58
- package/.bkit/core/workflows/bkit-build.md +967 -517
- package/.bkit/core/workflows/bkit-debug.md +555 -387
- package/.bkit/core/workflows/bkit-design.md +319 -283
- package/.bkit/core/workflows/bkit-plan.md +319 -342
- package/.bkit/core/workflows/bkit-ship.md +525 -328
- package/.bkit/shared/.agents/plugins/bkit/hooks.json +10 -0
- package/.bkit/shared/.bkit/.env.example +2 -2
- package/.bkit/shared/.claude/bkit-cloud.md +12 -10
- package/.bkit/shared/.claude/settings.json +14 -0
- package/.bkit/shared/.codex/bkit-cloud.md +98 -0
- package/.bkit/shared/.codex/hooks.json +20 -0
- package/.bkit/shared/.cursor/bkit-cloud.md +3 -3
- package/.bkit/shared/.cursor/mcp.json +1 -1
- package/.bkit/shared/.cursor/rules/bkit-rules.mdc +1 -1
- package/.bkit/shared/.gitattributes +1 -1
- package/.bkit/shared/.mcp.json +1 -1
- package/.bkit/shared/.opencode/opencode.json +12 -1
- package/.bkit/shared/CLAUDE.md +6 -2
- package/.bkit/shared/GEMINI.md +5 -2
- package/.bkit/shared/docs/architecture.md +4 -7
- package/.bkit/shared/docs/data-schema.md +4 -7
- package/.bkit/shared/docs/deployments.md +241 -168
- package/.bkit/shared/docs/design.md +11 -12
- package/.bkit/shared/docs/noise.md +8 -7
- package/.bkit/shared/docs/observability.md +5 -8
- package/.bkit/shared/docs/security.md +35 -17
- package/.bkit/shared/docs/setup.md +29 -18
- package/.bkit/shared/docs/technology.md +3 -6
- package/.bkit/shared/docs/testing.md +289 -174
- package/.bkit/shared/gitignore +6 -2
- package/.bkit/templates/.claude/hooks/bkit-cloud-bootstrap.sh +15 -16
- package/.bkit/templates/.cursor/Dockerfile +12 -8
- package/.bkit/templates/.cursor/environment.json +1 -1
- package/.bkit/templates/.github/ISSUE_TEMPLATE/bkit-bug.yml +10 -8
- package/.bkit/templates/.github/ISSUE_TEMPLATE/bkit-change.yml +45 -12
- package/.bkit/templates/.github/ISSUE_TEMPLATE/bkit-idea.yml +3 -3
- package/.bkit/templates/.github/renovate.json5 +34 -6
- package/.bkit/templates/.github/workflows/bkit-ci.yml +54 -51
- package/.bkit/templates/.github/workflows/bkit-cloud.yml +34 -33
- package/.bkit/templates/.github/workflows/bkit-health.yml +97 -0
- package/.bkit/templates/.github/workflows/bkit-renovate.yml +5 -4
- package/.bkit/templates/.github/workflows/bkit-trivy.yml +5 -4
- package/.bkit/templates/.husky/pre-commit +2 -2
- package/.bkit/templates/.husky/pre-push +96 -36
- package/.bkit/templates/.prettierignore +6 -0
- package/.bkit/templates/.vale.ini +27 -27
- package/.bkit/templates/context/bkit-cloudrun.md +23 -15
- package/.bkit/templates/context/bkit-expo.md +70 -52
- package/.bkit/templates/context/bkit-github.md +93 -56
- package/.bkit/templates/context/bkit-inngest.md +21 -15
- package/.bkit/templates/context/bkit-maestro.md +22 -16
- package/.bkit/templates/context/bkit-playwright.md +23 -17
- package/.bkit/templates/context/bkit-resend.md +12 -10
- package/.bkit/templates/context/bkit-sentry.md +30 -20
- package/.bkit/templates/context/bkit-serena.md +8 -7
- package/.bkit/templates/context/bkit-stripe.md +14 -13
- package/.bkit/templates/context/bkit-supabase.md +16 -13
- package/.bkit/templates/context/bkit-tavily.md +14 -11
- package/.bkit/templates/context/bkit-vercel.md +27 -18
- package/.bkit/templates/docs/capabilities.md +15 -14
- package/.bkit/templates/scripts/ci/{check-accepted-divergences.mjs → bkit-check-accepted-divergences.mjs} +6 -3
- package/.bkit/templates/scripts/ci/{check-accepted-divergences.test.mjs → bkit-check-accepted-divergences.test.mjs} +3 -3
- package/.bkit/templates/scripts/ci/{check-capstone.mjs → bkit-check-capstone.mjs} +2 -2
- package/.bkit/templates/scripts/ci/{check-capstone.test.mjs → bkit-check-capstone.test.mjs} +4 -4
- package/.bkit/templates/scripts/ci/{check-dependency-audit.mjs → bkit-check-dependency-audit.mjs} +103 -18
- package/.bkit/templates/scripts/ci/{check-dependency-audit.test.mjs → bkit-check-dependency-audit.test.mjs} +31 -3
- package/.bkit/templates/scripts/ci/bkit-machine-lock.sh +387 -0
- package/.bkit/templates/scripts/ci/bkit-test-awake.mjs +75 -0
- package/.bkit/templates/scripts/ci/bkit-test-awake.test.mjs +160 -0
- package/.bkit/templates/scripts/ci/lanes/{resolve-push-base.mjs → bkit-resolve-push-base.mjs} +3 -3
- package/.bkit/templates/scripts/ci/lanes/{resolve-push-base.test.mjs → bkit-resolve-push-base.test.mjs} +4 -4
- package/.bkit/templates/scripts/cleanup/bkit-disk-prune.mjs +10 -2
- package/.bkit/templates/scripts/graphify/bkit-graph-drift.py +1 -1
- package/.bkit/templates/scripts/graphify/bkit-graph-eager.py +3 -3
- package/.bkit/templates/scripts/graphify/bkit-graph-embed.py +63 -0
- package/.bkit/templates/scripts/graphify/bkit-graph-imports.py +2 -2
- package/.bkit/templates/scripts/graphify/bkit-graph-pre-push.sh +69 -18
- package/.bkit/templates/scripts/graphify/bkit-graph-repair.py +3 -3
- package/.bkit/templates/scripts/graphify/bkit-graph-surface.py +1 -1
- package/.bkit/templates/scripts/graphify/bkit-graphify-graphml.py +1 -1
- package/.bkit/templates/scripts/graphify/bkit-graphify-rollup.py +197 -126
- package/.bkit/templates/scripts/graphify/bkit-graphify.sh +64 -41
- package/.bkit/templates/scripts/slop/bkit-slop-type-suppression.mjs +23 -7
- package/.bkit/templates/slop-scan.config.json +9 -2
- package/CHANGELOG.md +231 -0
- package/README.md +278 -89
- package/dist/checks/after/auth.js +36 -33
- package/dist/checks/after/ci-yml.js +91 -30
- package/dist/checks/after/github.js +91 -17
- package/dist/checks/after/index.js +22 -11
- package/dist/checks/after/suppression.js +8 -5
- package/dist/checks/after/tools.js +111 -114
- package/dist/checks/before/context-fetch.js +218 -33
- package/dist/checks/before/env.js +3 -3
- package/dist/checks/before/files.js +5 -3
- package/dist/checks/before/index.js +20 -3
- package/dist/checks/before/repo.js +197 -2
- package/dist/checks/status-holes.js +739 -114
- package/dist/checks/tips/catalogue.js +70 -19
- package/dist/checks/tips/detect.js +62 -0
- package/dist/checks/tips/index.js +51 -9
- package/dist/cli.js +234 -124
- package/dist/commands/analysis.js +174 -141
- package/dist/commands/ci/ci.js +302 -25
- package/dist/commands/ci/pins.js +142 -0
- package/dist/commands/ci/push-gate.js +194 -0
- package/dist/commands/comply.js +54 -91
- package/dist/commands/install/doctor.js +2 -2
- package/dist/commands/install/install.js +50 -19
- package/dist/commands/install/uninstall.js +82 -55
- package/dist/commands/install/upgrade.js +21 -11
- package/dist/commands/land.js +137 -17
- package/dist/commands/lib.js +295 -21
- package/dist/commands/registry.js +10 -2
- package/dist/commands/setup/setup.js +8 -7
- package/dist/commands/stats.js +175 -0
- package/dist/commands/tips.js +7 -3
- package/dist/commands/workflow/branches.js +1873 -152
- package/dist/commands/workflow/challenge.js +609 -0
- package/dist/commands/workflow/completeness.js +97 -0
- package/dist/commands/workflow/context.js +1045 -46
- package/dist/commands/workflow/counters.js +385 -86
- package/dist/commands/workflow/decisions.js +484 -0
- package/dist/commands/workflow/disk.js +214 -56
- package/dist/commands/workflow/fids.js +706 -33
- package/dist/commands/workflow/issues.js +349 -59
- package/dist/commands/workflow/ledgers.js +1123 -37
- package/dist/commands/workflow/locks.js +26 -2
- package/dist/commands/workflow/plan-blocks.js +501 -0
- package/dist/commands/workflow/plan-versions.js +22 -0
- package/dist/commands/workflow/plans.js +205 -0
- package/dist/commands/workflow/prs.js +154 -35
- package/dist/commands/workflow/repo.js +58 -8
- package/dist/commands/workflow/roadmap.js +58 -23
- package/dist/commands/workflow/run-facts.js +421 -0
- package/dist/commands/workflow/runs.js +670 -72
- package/dist/commands/workflow/runtime.js +15 -1
- package/dist/commands/workflow/skills.js +1 -0
- package/dist/commands/workflow/status.js +282 -34
- package/dist/commands/workflow/step-table.js +308 -0
- package/dist/commands/workflow/steps.js +703 -96
- package/dist/commands/workflow/summary-findings.js +117 -0
- package/dist/commands/workflow/summary.js +388 -0
- package/dist/commands/workflow/tasks.js +378 -0
- package/dist/commands/workflow/test-blocks.js +231 -0
- package/dist/commands/workflow/tests-changed.js +172 -0
- package/dist/commands/workflow/tests.js +1695 -305
- package/dist/commands/workflow/tracker.js +983 -0
- package/dist/commands/workflow/workflows.js +1014 -222
- package/dist/commands/workflow/worktrees.js +432 -26
- package/dist/install/cloud-env.js +233 -99
- package/dist/install/context-ci-workflow.js +85 -0
- package/dist/install/manifest.js +64 -17
- package/dist/install/modules/antigravity.js +3 -2
- package/dist/install/modules/cloud.js +25 -16
- package/dist/install/modules/codex.js +12 -2
- package/dist/install/modules/core.js +1090 -48
- package/dist/install/modules/cursor.js +6 -6
- package/dist/install/modules/opencode.js +13 -7
- package/dist/install/template-notes.js +116 -0
- package/dist/install/version-file.js +9 -4
- package/dist/ledgers/compact.js +87 -0
- package/dist/ledgers/journal.js +233 -0
- package/dist/ledgers/migrate.js +81 -0
- package/dist/ledgers/status.js +176 -13
- package/dist/ledgers/telemetry.js +1010 -125
- package/dist/render/hosts.js +104 -14
- package/dist/render/models.js +12 -0
- package/dist/shared/bkit-latest.js +26 -4
- package/dist/shared/branches.js +3 -1
- package/dist/shared/dep-updater.js +49 -0
- package/dist/shared/egress.js +8 -1
- package/dist/shared/json.js +20 -7
- package/dist/shared/package-manager.js +6 -0
- package/dist/shared/settings.js +181 -24
- package/dist/shared/text.js +333 -0
- package/dist/shared/timeouts.js +2 -0
- package/dist/tools/cli-doc.js +97 -0
- package/dist/tools/grammar-density.js +183 -0
- package/dist/tools/instructions-sources.js +720 -0
- package/dist/tools/instructions.js +1623 -0
- package/package.json +156 -51
- package/.bkit/core/.bkit/lite.md +0 -73
- package/.bkit/core/engine/commands/harvest.ts +0 -459
- package/.bkit/core/engine/commands/workflow/workstreams.ts +0 -166
- package/.bkit/core/engine/tools/wording.ts +0 -748
- package/.bkit/core/prompts/bkit-build-challenge-reconcile.md +0 -69
- package/.bkit/core/prompts/bkit-build-challenge-remediate.md +0 -71
- package/.bkit/core/prompts/bkit-build-challenge-tests.md +0 -74
- package/.bkit/core/prompts/bkit-build-plan-execute.md +0 -76
- package/.bkit/core/prompts/bkit-build-summary.md +0 -94
- package/.bkit/core/prompts/bkit-debug-summary.md +0 -96
- package/.bkit/core/prompts/bkit-design-summary.md +0 -107
- package/.bkit/core/prompts/bkit-plan-complexity-check.md +0 -67
- package/.bkit/core/prompts/bkit-plan-complexity-split.md +0 -88
- package/.bkit/core/prompts/bkit-plan-review-3.md +0 -93
- package/.bkit/core/prompts/bkit-plan-scope-check-1.md +0 -85
- package/.bkit/core/prompts/bkit-plan-scope-check-2.md +0 -85
- package/.bkit/core/prompts/bkit-plan-summary.md +0 -107
- package/.bkit/core/prompts/bkit-ship-summary.md +0 -78
- package/.bkit/templates/scripts/ci/bkit-local-stack-lock.sh +0 -194
- package/dist/commands/harvest.js +0 -353
- package/dist/commands/workflow/workstreams.js +0 -143
- package/dist/tools/wording.js +0 -680
- /package/.bkit/templates/.vale/styles/{House → Bkit}/Filler.yml +0 -0
- /package/.bkit/templates/.vale/styles/{House → Bkit}/Hedging.yml +0 -0
- /package/.bkit/templates/.vale/styles/{House → Bkit}/LLMSlop.yml +0 -0
- /package/.bkit/templates/.vale/styles/{House → Bkit}/Verbose.yml +0 -0
- /package/.bkit/templates/scripts/ci/{accepted-advisories.json → bkit-accepted-advisories.json} +0 -0
- /package/.bkit/templates/scripts/ci/{accepted-divergences.json → bkit-accepted-divergences.json} +0 -0
- /package/.bkit/templates/scripts/launchd/{launchd.env.example → bkit-launchd.env.example} +0 -0
- /package/.bkit/templates/scripts/launchd/{ai.bkit.launchd.plist → bkit-launchd.plist} +0 -0
package/CHANGELOG.md
CHANGED
|
@@ -6,6 +6,237 @@ Format: `## [version] — YYYY-MM-DD`
|
|
|
6
6
|
|
|
7
7
|
---
|
|
8
8
|
|
|
9
|
+
## [6.1.0] — 2026-11-04
|
|
10
|
+
|
|
11
|
+
### Breaking Changes
|
|
12
|
+
|
|
13
|
+
- **BREAKING — Every name a consumer types or sees now carries the bkit brand.** Package scripts: `lint:prose` → `bkit:lint:prose`, `slop:delta` → `bkit:slop:delta`, `graph*` → `bkit:graph*`. Files: `scripts/launchd/ai.bkit.launchd.plist` → `bkit-launchd.plist`, `launchd.env.example` → `bkit-launchd.env.example`, and the CI scripts under `scripts/ci/` (`check-capstone`, `check-dependency-audit`, `check-accepted-divergences`, `lanes/resolve-push-base`, `accepted-advisories.json`) gain `bkit-`. Check names: every workflow `name:` and job display name in the shipped workflows starts with `bkit-` (`bkit-renovate`, `bkit-trivy` and the cloud jobs included). Vale style: `House` → `Bkit`, so `.vale.ini` names `Bkit.*`. `bkit upgrade` migrates all of it: it renames the keys in `package.json`, the tokens in `.husky/pre-commit`, `.husky/pre-push` and the `bkit.testing.ciWorkflow` file, moves unedited CI script copies and Vale rules that still match the shipped text, rewrites shipped workflow names, and leaves anything you edited alone; a second run changes nothing. Branch protection: a required status check that names an old job (for example `Renovate` or `Trivy`) no longer matches; update the required checks in the repo settings to the `bkit-*` names after upgrading. Generic stack scripts (`typecheck`, `lint`, `format:check`, `test`, `build`, `lint:slop`, `prepare`) keep their names. #748
|
|
14
|
+
|
|
15
|
+
### Added
|
|
16
|
+
|
|
17
|
+
- **Local tests and gates keep a Mac awake while they run.** Engine test commands hold a `caffeinate -is` assertion, and the copy-and-own `scripts/ci/bkit-test-awake.mjs` wrapper covers direct runner commands and complete gate chains. Cancellation reaches the test process and releases the assertion. Other platforms run the original command, and test timeouts stay unchanged. See `.bkit/templates.md` § Local test sleep prevention.
|
|
18
|
+
- **Workflow steps show an engine-derived progress line.** Each numbered completion returns an exact ten-cell bar, percentage and step title for the agent to show as plain text. Build lines reflect current task groups, retries show the saved attempt allowance, and resuming recomputes the same position from the run's status and plan. Explicit gate-only completions use the same format without adding status rows. `bkit upgrade` installs the narration on every supported host.
|
|
19
|
+
- **`bkit branches push` hands pre-push tests-lock waits back to the agent.** It passes the host's lock wait budget and its run name to Git, and exits 81 with `lockHolder` when the hook's gate gives up waiting. Build, debug and ship re-run their push blocks on 81.
|
|
20
|
+
|
|
21
|
+
- **`bkit tests changed` recognizes the same tests-lock outcomes as `tests run`.** It accepts `--lock-wait-s`, passes the budget and its run name to the selected gate, and exits 81 with `lockHolder` when that gate gives up on the tests lock. Both commands keep a wait separate from a failing test run; `tests run --file` appends no `test-run` status row for a wait. The caller prompts re-run the command on exit 81.
|
|
22
|
+
- **`bkit tests run` hands tests-lock waits back to the agent.** It passes `--lock-wait-s` and its run name to a wrapped gate, and exits 81 with `lockHolder` when that gate gives up on the tests lock. A refusal exits 4; gate failures keep their normal verdict. The wait defaults to `BKIT_LOCK_WAIT_S`, else 60 seconds; Claude Code ships a 1200-second budget. Light test-first runs stay unwrapped.
|
|
23
|
+
- **The machine lock queues test gates across repositories.** The template is renamed `bkit-machine-lock.sh` and adds named machine-wide locks, configurable test slots, worker limits and enforced lock order. `run` exits 73 when a lock cannot be taken and 75 when its wait gives up; without a TTY, the default wait is 60 seconds. Adopted stack-lock copies keep working. To adopt the tests lock, copy `.bkit/templates/scripts/ci/bkit-machine-lock.sh` to `scripts/ci/` and wrap each test-running `bkit:gate:*` script with `bash scripts/ci/bkit-machine-lock.sh run --lock tests -- <gate command>`. Put chained commands in one `sh -c` string. See `.bkit/templates.md` § Machine lock.
|
|
24
|
+
- **A setup tip tells you when a newer bkit is out.** `bkit tips` shows `bkit-upgrade` first, ahead of every other tip in every workflow, while the newest release `bkit repo info` last saw (cached once a day in git's common dir) is newer than `.bkit/version`. The tip only reads that cache and never asks the registry itself, so it stays quiet until a workflow has run once. It cannot be silenced: `bkit.advisories.ignore` skips it, a workflow's silence line never names it, and `BKIT_NO_UPDATE_CHECK=1` stops only the lookup, not a release already seen. Its ask points the agent at `.bkit/install.md` § Upgrade — already installed.
|
|
25
|
+
- **A setup tip offers to send your bkit friction and stats to bkit's makers as one issue.** `bkit tips` shows `share-friction` while `.bkit/friction.jsonl` holds an `OPEN` entry that is not `consumer` scope. Its paste block, `.bkit/README.md` § Share friction with bkit, has the agent refresh `.bkit/stats.md`, strip each entry to the fields bkit needs, remove anything that names the project, show you the exact issue before posting, and then mark each entry `TRIAGED` with the issue link, so the tip returns only when new friction arrives. It never fires in bkit's own repo.
|
|
26
|
+
- **`bkit ci bookkeeping-push` tells the pre-push hook when a `[skip ci]` push to an open PR is safe.** Ship's Step 8 pushes its docs, logs, ledgers and skill removals as one `[skip ci]` commit onto the PR, and the merge gate then checks CI at the last tested head (validated-head carry-forward). The hook's CI-suppressor guard blocked exactly that push, so every ship run stopped at Step 8. The hook now hands the head the remote holds to `bkit ci bookkeeping-push --from=<sha>`, which exits 0 when every file changed since that head is on ship's `--scope=pr` allow-list, and lets the push through with a line saying why; a rewritten head, a code file, or no remote head still blocks. Both hook copies carry the change. Your `.husky/pre-push` is your own copy and `bkit upgrade` never touches it: paste the `CI-suppressor HEAD guard` block and the `PUSH_BRANCH_REMOTE_SHA` lines of the ref loop from `.bkit/templates/.husky/pre-push`. #699
|
|
27
|
+
- **`bkit.testing.ciSkipPaths` names the project docs ship's `[skip ci]` bookkeeping commit may carry.** Ship's `--scope=pr` allow-list reaches top-level `docs/*.md` and nothing else outside the frozen list; a repo that keeps project docs elsewhere lists them in `package.json` under `bkit.testing.ciSkipPaths` (globs as in `minPushPaths`: `*` one folder, `**` any depth; install seeds `[]`). bkit's own `package.json` lists `CHANGELOG.md` and `README.md`, the two ship-phase reads in `.agents/docs.md`; the engine hardcodes no file name. One list feeds the Step 8 commit, `bkit ledgers only-bookkeeping`, the merge gate's carry-forward and the hook's exception, so a changelog edit the docs sync makes no longer stays behind uncommitted in the worktree. #699
|
|
28
|
+
- **A separate subagent writes each task group's tests before another writes its code.** `/bkit-build` Step 7 now runs two subagents per group: `@bkit-build-plan-tests` writes the group's planned `TEST FIRST` tests and may touch nothing else (`bkit steps finish --step=build-plan-tests --plan` refuses a diff that changes any other file, exit 15), the orchestrator records every test failing with one `bkit tests red --group` run, and only then does `@bkit-build-plan-tasks` write the code. The code steps may not change a planned test: `steps finish` on execute and gaps-remediate exits 11 when a planned test file differs from the blob its red row recorded. `bkit context brief --tests-only` writes the tests writer's brief (`build-brief-tests-{GROUP}-{n}.md`), and with `--file` leaves out the tests the group's red rows already proved, so a handover or a task added at a Decision Pause briefs only what is left; `tests red` skips those tests too while their file is unchanged. Three in-flight builds had written code before any red with one agent doing both, however the prompt ordered it; separating the writers makes the order a property of the workflow, not of the agent. Plan caps rise with it: `bkit.workflows.tasks.maxPerGroup` defaults to 5 (was 3) and `maxTestsPerGroup` to 25 (was 15), so the number of subagents per plan stays about the same; a value you set is never touched, and `bkit check` names the number to change in `.github/ISSUE_TEMPLATE/bkit-change.yml`. `bkit upgrade` installs the new agent and workflow. #796
|
|
29
|
+
- **The pre-push hook writes `tmp/pre-push.log`.** `.bkit/templates/.husky/pre-push` now runs itself once more with everything it prints copied to `tmp/pre-push.log`, rewritten each push, and prints that path as its first and last line with the exit code. An agent whose command output was cut short, or that pushed in the background, can read the whole run there; a second terminal can `tail -f` it while the push runs. The exit code and the ref list git passes the hook are unchanged. `git push --no-verify`, and a checkout without husky installed, run no hook and write no log. Your `.husky/pre-push` is your own copy and `bkit upgrade` never touches it: to adopt this, paste the `# Log:` block from `.bkit/templates/.husky/pre-push` right after `set -e`. #742
|
|
30
|
+
- **`bkit ledgers compact` shrinks the five keyed ledgers to one row per entry, without losing an entry's age or how often it changed.** `docs/knowledge.jsonl`, `docs/decisions.jsonl`, `docs/deferrals.jsonl`, `docs/failures.jsonl` and `.bkit/issues.jsonl` keep growing because every update is a new row. The verb drops each row a newer row of the same entry replaces, and leaves the surviving row where it was. That row gains `firstTs` (when the entry was first written) and `updates` (how many rows it has had), so age reports and flake counts still work. Every reader sees the same entries as before. Lines it cannot read, and rows of other kinds, are left exactly as they are. Running it twice changes nothing, and if a merge brings dropped rows back, the next run drops them again. `--ledger=<name>` limits it to one ledger, and `--dry-run` only counts. `bkit ledgers append` and `bkit roadmap feature` now fill `firstTs` / `updates` on every new row of those ledgers. Each of the five ledger updaters (`@bkit-utility-update-knowledge`, `-decisions`, `-deferrals`, `-failures`, `-roadmap`) ends by compacting its own ledger. No upgrade step is needed: rows without the two fields read as entries with one row.
|
|
31
|
+
- **`.bkit/cli.md` lists every `bkit` command in one place.** The commands you run (`bkit install`, `bkit upgrade`, `bkit check` and the rest) and every `bkit <noun> <verb>` command the workflows run, one row per verb with its flags and whether it is safe to run by hand or only as a workflow step. The command contract (output format, flag names, shared exit codes, `{BKIT}` invocation, step endings, `{OUTBOX}`) moves there from `.bkit/workflows.md` § Workflow commands, which now points at it. The tables are generated from the same specs `--help` prints, so they cannot drift from the code. `bkit upgrade` adds the file. #741
|
|
32
|
+
- **A ledger or lockfile conflict no longer aborts ship's integration when another conflict needs judgement.** `bkit branches sync --merge` (ship Step 4) resolves the append-only ledgers and the lockfile, leaves the other paths conflicted with the merge still in progress, lists them under `pending`, and exits 24 with outcome `pending`. Run it again once those paths are staged by hand, and it commits the merge. `@bkit-ship-conflict-resolve` re-runs the merge block after its hand resolution, so a merge is no longer aborted when its only other conflict is one the agent may resolve by hand. `bkit upgrade` picks it up. #747
|
|
33
|
+
- **`bkit completeness check` tells you which steps and bkit commands a workflow run skipped.** It reads a run's `status.jsonl` and the workflow markdown it ran (design, plan, build, debug or ship), and reports the steps the markdown requires that have no row, plus the `bkit` commands a step names that the run never made. The expected set comes from the markdown at run time, so there is no table to keep in step. It is advisory: it always exits 0 and never blocks a run. To make the commands visible, every `bkit` command now appends one `engine-call` row (noun, verb, flag names) to the active run's `status.jsonl`, and a command that exits non-zero for a reason the workflow does not route on appends one `tool-failure` friction row to `.bkit/friction.jsonl`. `BKIT_ENGINE_JOURNAL=off` turns both off. `bkit workflows close` writes the verdict as a `completeness` row, and `bkit workflows start` / `resume` prints a warning when the FID's previous run closed incomplete. The CI static lane's `pnpm bkit:audit:status` sweep now judges every run in `docs/plans/*/logs/*/status.jsonl` this way and still never fails the lane on a hole. A workflow step a run may skip is marked with a column-0 `**Run only if**` or `**Skip if**` line, and every other step writes a row on every route (`.bkit/authoring.md`); `/bkit-build` Step 4 (Plan Convert) now writes a skip row when the plan is already in task-block form. `bkit upgrade` picks it up. #699
|
|
34
|
+
- **Answers to a design's or plan's open questions reach the document, the issue and the ledger.** After `/bkit-design` or `/bkit-plan` closes, the orchestrator records every round of answers the owner gives to the summary's `## Open Questions`, whether through the one-by-one pass or in free chat. Each round appends a `docs/decisions.jsonl` row for every answer that settles an architecture or technology decision (as before), writes the questions and answers verbatim to `docs/plans/{YYYYMM}/logs/{FID}/design-answers-{RUN}.md` (plan: `plan-answers-{RUN}.md`), has `@bkit-design-update` / `@bkit-plan-update` edit the canonical `{FID}-design.md` / `{FID}-plan.md` in place (the answered question leaves the open questions, the decision is written down, and statements the answer overrides follow it; a plan is re-checked with `bkit tests plan-check`), commits those files on `dev` without pushing, and, with a tracker connected, posts the round as one `Owner Answers` comment on the issue. Before this, only the decision row was written, so the next workflow read the question as still open unless the owner edited the document by hand. `bkit upgrade` picks it up. #798
|
|
35
|
+
|
|
36
|
+
### Changed
|
|
37
|
+
|
|
38
|
+
- **Permanent skills are checked for each configured host, and activate for the task that needs them.** The registry checks the skill file, installer lock and host discovery together. A skill's demand to activate in every session does not override the owner's request, the repository's workflow or its test rules. Missing installations are repaired through the skills CLI, while owner approvals and existing work are preserved.
|
|
39
|
+
- **Bookkeeping increments can reuse a successful local full gate.** The pre-push hook sends every outgoing old/new commit pair to `bkit ci min-push`. It selects `bkit:gate:min` when the changes after a locally checked ancestor contain only allowed bookkeeping files and the recorded script definitions still match. Missing or invalid evidence, a failed full check, rewritten history or new code keeps the full gate. This does not let code skip CI.
|
|
40
|
+
- **Workflow steps have no universal runtime deadline.** Attempt limits, subagent turn caps, command wait budgets and host execution limits still apply.
|
|
41
|
+
- **Prompts and workflows name a section by its bare title, and keep the `#` marks only for a line the agent writes or matches.** `.bkit/authoring.md` § All Agent-Read Files adds the **Section names** rule: "the design's `Acceptance Scenario`" names a section to read, "append `## Project Instructions`" names a line to write, and a fixed line the agent recognises a file by (the `## Owner answers` heading) or a program prints or parses (`## bkit Tips`, `###### bkit:blocked-by`) keeps its marks. The challenge, plan-convert and plan prompts, the build, design and plan workflows, and the shared prompt and workflow docs now follow it.
|
|
42
|
+
- **`bkit.testing.shipGate` is `bkit.testing.ciRequired`, and `bkit upgrade` renames it for you.** The list of CI lanes whose `skipped` conclusion refuses a ship merge now sits with the other CI keys as `ciRequired`, in the order `minPushPaths`, `ciWorkflow`, `ciCapstone`, `ciRequired`, `ciSkipPaths`. A `testing.shipGate` in `package.json` moves to `testing.ciRequired` on the next install or upgrade; a `ciRequired` already set wins. `bkit runs gate` reads only the new key. #699
|
|
43
|
+
- **A min push is what bkit writes plus what `bkit.testing.minPushPaths` adds, so the setting no longer restates bkit's own files.** `bkit ci min-push` now starts from ship's `--scope=pr` bookkeeping allow-list (ledgers, logs, `docs/*.md`, skill removals, the repo's `bkit.testing.ciSkipPaths`) plus `docs/plans/**`, the same list the ship gate and the pre-push `[skip ci]` guard accept untested, and `minPushPaths` only adds to it; `[]` means that floor alone, not off. Installs seed `[]` instead of `docs/plans/**`, `docs/*.jsonl`, `.bkit/*.jsonl`, and `bkit upgrade` drops those three and anything already in `ciSkipPaths` from a repo's list. #699
|
|
44
|
+
- **BREAKING — The subagent that builds a group's tasks is `@bkit-build-plan-tasks`.** `.bkit/core/prompts/bkit-build-plan-tasks.md` replaces `bkit-build-plan-execute.md`, and every host mirror, workflow line, doc and instruction entry names it; `bkit upgrade` writes the new agent to each host and deletes the old mirror files. The status step is `build-plan-tasks` too (`steps finish --step=build-plan-tasks`, `build-plan-tasks-{GROUP}-log.md`, verdict `TASKS_VERDICT`); Step 7 keeps its title. `bkit upgrade` renames a consumer's `## Model Overrides` row for `bkit-build-plan-execute` to the new agent, provider and model kept. A build in flight across the rename resumes: a `status.jsonl` row or group log under the old step name is read as the new one. Ledger rows and logs written before the rename keep the old name. #796
|
|
45
|
+
- **A build locks each planned test case, not its file, and asks about unproven tests once, at the end.** `bkit tests red` records each planned test's own lines (`block`, a hash found by the test's name and indentation), and `bkit steps finish` refuses a code step (exit 11) only when one of those tests changed, naming each in `tests`; other tests, helpers and new tests in the same file are the code step's to change, and every refusal appends an `agent-rework` friction row. An existing test the tests step updated and that passes is recorded `updated`, and the verifier reads its diff: it may add expectations, never remove or loosen one. A planned test with no proven red is no longer a `no-red` gap that stops the run mid-way: `bkit tests green` lists it under `unproven`, and `/bkit-build` Step 15 asks the owner once about all of them, Accept writing a `TEST EXCEPTION`. `bkit tests plan-check` reports planned test names already in use (`existing`). Rows written by an older bkit lock by the whole file as before. #801
|
|
46
|
+
- **A used-up fix loop asks the owner instead of failing the run.** With `ask` true and not headless, a used-up `build.verify`, `build.tests`, `debug.soak` or `ship.merge` limit goes to a question; a go-on answer runs the new `bkit counters grant`, which writes an `allowance-granted` row giving that loop alone (that group alone for verify) one more full allowance of `maxAttempts`. `counters route` and `counters max` print `grants` and `allowance`, and `challenge merge --round` accepts the larger limit. With `ask` false or headless, a used-up limit still goes to the Failure Path. #801
|
|
47
|
+
- **BREAKING — Test-first is judged per group from one red row.** `bkit tests red` takes `--group` and `--plan` and runs every planned test file of the group's open tasks once, writing one `build-tests-red` row with a state per test (`red`, `red-not-loaded`, `passed-accepted`, `vacuous`, `missing`, `no-report`, `lint`); the per-test flags `--task`, `--test` and `--name`, the `build-task-red` and `build-task-red-passed` rows, the `.snapshot` copies and the per-test logs are gone. A test file that fails to load counts as red: the tests step writes no source, so an import of unwritten code is the expected shape. `bkit tests green` reads the group's red rows, merged per test, and no longer judges test order or test edits: the `late-red`, `red-mismatch`, `red-evidence-missing` and `test-changed` gaps are gone, because the code step cannot change a planned test (its finish refuses the step) and the red row is written before it runs. `build-tests-red-*.md` replaces `build-task-red-*.md` in `.prettierignore`; builds in flight on the old rows finish on the old bkit. #796
|
|
48
|
+
- **BREAKING — `bkit.dependencies.updater` records the updater you run, and `bkit check` holds it to the repo.** The value is `"renovate"`, `"dependabot"` or `"none"`. `bkit install` / `bkit upgrade` fill an empty value from the one updater config present. `bkit check` now blocks on `dep-updater-unset` (a config is present but the value does not name it) and `dep-updater-mismatch` (the value names the other updater, or `none`, while a config is present; or names an updater with no config). Renovate config is found at every location Renovate reads, not only `.github/renovate.json5` / `.json`: root `renovate.json*`, `.gitlab/renovate.json*`, `.renovaterc*` and a `package.json` `renovate` block. Run `bkit upgrade` to have the value set; a repo running both updaters, or a value set by hand that disagrees with its config, needs a hand fix.
|
|
49
|
+
- **opencode columns measure every pick against Claude, and the four key gates run `gpt-6.1-sol`.** On OpenRouter, Ollama, OpenCode Zen and OpenCode Go each agent now gets the cheapest model that comes closest to the Claude model it runs on, and any vendor may be picked; the weekly cost cap, not a vendor list, keeps the dearest models off. Four checks where a miss reaches the merge unnoticed — `bkit-build-plan-verify`, `bkit-ship-deferred-sweep`, `bkit-debug-failure-diagnose` and `bkit-build-challenge-2` — move to `gpt-6.1-sol` on OpenRouter and Zen (68.7 on LiveBench 2.0 Long Agentic) and `gemini-4-argon` on Go, which serves no Sol; every other agent, and every Ollama agent, runs DeepSeek V4.1 Flash. `mimo-v2.6-pro`, `minimax-m3.1-flash` and `glm-5.3-flash` leave the judge cells. A two-day OpenRouter overrun on the default comes to about A$74 against the A$70 weekly cap; Option 2 in § Model Presets puts five more judges on Sol for about A$137. Effort follows the same yardstick: DeepSeek thinks at its maximum everywhere, while `gpt-6.1-sol` (and any Claude or `gpt-6-astra` override) keeps the agent's own Claude effort instead of being raised to the top rung. `bkit upgrade` picks it up. #205 #714
|
|
50
|
+
- **The opencode OpenRouter and Ollama columns pin the dated DeepSeek V4.1 Flash `1017` snapshot.** The undated `deepseek/deepseek-v4.1-flash` slug OpenRouter serves is frozen on the 2026-09-10 snapshot and does not float, so the two columns move to the October 17 refresh: `deepseek/deepseek-v4.1-flash-1017` (35 endpoints) on OpenRouter and `deepseek-v4.1-flash:1017` on Ollama. DeepSeek's model card puts 1017 ahead of the previous snapshot across its agentic benchmarks (Terminal-Bench 4.0 31.2 → 42.3, Apex Bench v1.1 64.6 → 72.1) with shorter outputs and faster task completion, and reports no API price change; the OpenCode Zen and Go gateways carry no dated snapshot yet and stay on the undated slug. `bkit upgrade` picks it up. #205
|
|
51
|
+
- **A `/bkit-build` run's step comments carry their own numbers, and build has 16 steps.** Plan Convert is its own Step 4, posting `### Step 4 — Plan Convert` instead of sharing Branch Sync's number and heading; Branch Sync keeps Step 3 with `bkit branches sync --merge-docs` alone. Plan Current moves to Step 5 and posts on both routes — `Status: skipped (no plan anchor)` when `bkit branches anchor-read` exits 12, the agent's comment otherwise — so a converted plan no longer leaves the step silent. Every later step shifts by one, through Summary 15 and Close Workflow & Create PR 16, and `WORKFLOW_STEPS` with its resume targets moves with them. A plan already valid with no newer comments writes no Plan Convert row and posts no comment. A run that started on the old numbering finishes on the bkit version it started on. #705
|
|
52
|
+
- **The opening note of every doc template now sits inside the managed block, so `bkit upgrade` keeps it current.** The `<!-- Note: … -->` line used to sit above `<!-- bkit:start -->`, where upgrade never looks, so an installed `docs/templates/bkit-*.md` kept whatever note it was first installed with. It now opens the block, right after the managed marker. `bkit upgrade` removes the old copy above the block when it matches a note wording bkit shipped there, along with the blank line after it. A note you edited is kept, and upgrade prints a line naming the file when the note still names a bkit report or plan path or a `@bkit-` agent. Every other line above the block stays, including your own `<!-- Note: … -->` comments, with no warning. A file with CRLF line endings throughout is left as it is. `bkit check` fails until upgrade has run. #747
|
|
53
|
+
- **A direct chat can write every ledger, not just `docs/decisions.jsonl`.** `bkit ledgers append` now accepts `utility` rows of every kind: decision, deferral, failure, knowledge, friction, compliance, counter and outcome. A fix made in a chat session can record its gotcha or close its deferral in the same change. Rows written this way carry `utility`, which keeps them apart from workflow runs in per-workflow reports; the overall totals still count them, which #766 fixes. #747
|
|
54
|
+
- **The three design ideate agents say which lens each one takes.** `bkit-design-brainstorm-ideate-1`, `-2` and `-3` shared one description and Role line, so an agent picker could not tell them apart. Each now names its lens: pragmatic, long-term or risk-tested.
|
|
55
|
+
- **The `Debug on demand` rule says when it applies.** In `.agents/rules.md` it now reads as the last step after `Local tests first` and `Existing evidence first`: re-run a failing E2E / `@live` test against the deployed system only once both have left it unexplained. What it asks for is unchanged.
|
|
56
|
+
- **Agents may dig deeper inside a step.** The `No improv` line every workflow and agent carries now reads "Skip no step, and add no step of your own; how deeply you investigate within a step is your call." It used to read as a ban on any extra investigation.
|
|
57
|
+
- **`Finish over deferrals` and `Scope discipline` no longer pull against each other.** `Finish over deferrals` now says what "when in doubt" covers: work the request needs before it is complete is in scope, and unrelated improvements are not.
|
|
58
|
+
- **`Research / investigate first` drops its fixed search queries.** It now asks you to check whether the framework or runtime already provides what you are about to build, and what current practice is, instead of prescribing two web searches for every change.
|
|
59
|
+
- **`Investigation layers` left `.agents/rules.md`.** It was coaching for one agent, loaded into every session. It now sits in `bkit-design-brainstorm-ideate-2`, the only agent that used it.
|
|
60
|
+
- **`Secrets sync`, `CI on PR to dev / main` and `CI on push to dev / main` apply only where the repo adopted the template they describe.** A repo without the secrets sync or CI workflow template no longer gets rules it cannot follow.
|
|
61
|
+
- **The logs, database and gates audits check that each scheduled job succeeded, not just that it ran.** A job that finishes fast with a rejected status used to pass all three. `bkit-audit-logs` Step 12 — Alarms, `bkit-audit-database` Step 12 — Derived Data and `bkit-audit-gates` Step 3 — Widen now share one sentence: read the outcome of every run (status code or exit code), and treat a job that did not write its owned rows in its window as a finding, even when it exited 0. The gates audit also lists a check on scheduled controls among the standard gates nobody proposed. #770
|
|
62
|
+
- **The gates and testing audits catch a test lane that goes green while its tests fail.** `bkit-audit-gates` Step 12 — Health compares each lane's runner-report failure count with the step's conclusion and rates `success` with failures `CRITICAL`; a lane with no machine-readable report goes to `Not Checked`. Its hollow-gate shapes add a trailing `; :` or `|| :`, `continue-on-error: true` on a test step, and a skip guard on a file that may be deleted (e.g., `if: hashFiles(...)`). `bkit-audit-testing` Step 7 — Shapes adds an exit status swallowed by a wrapper, and Step 9 — Runs counts test files the runner does not collect (e.g., `*.test.sh`). #771
|
|
63
|
+
- **The gates audit searches history for weakened gates and asks report-only gates for a count.** Step 11 — Re-Review also searches git history (e.g., `git log -S`) for gate steps that were removed or weakened, and a change with no `docs/decisions.jsonl` row is a finding. Step 9 — Sequence asks each report-only gate to print its finding count on every run, so Step 12 can compute the false-positive rate. #780
|
|
64
|
+
- **The testing audit can run hand mutants without editing the shared checkout, and knows two more timing shapes.** When in-place edits are blocked, the `Mutation Tool` fallback and Step 6 run the hand mutants in a throwaway worktree. Step 7 — Shapes adds absence asserted before it could appear (no wait after the trigger) and a listener attached after the event it observes. #784
|
|
65
|
+
- **The dependencies audit checks when the bot really runs, what the batch really resolves to, and checksum-paired pins.** Step 6 compares the bot's actual run start times with its configured window, because a window no run lands in means routine PRs never open. Step 8 records the resolved version of each changed package per workspace after the batch install, and checks each package-manager override against the direct pin it shadows. A batch that permissions stopped from assembling gets the new batch verdict `UNVERIFIED` instead of FAIL. Step 7 adds class **C, checksum** for a pin paired with a checksum, checked even when the PR is green. Step 10 re-runs a timeout under heavy load (e.g., exit 143) alone before blaming the batch. The anchor standards read package-manager defaults from the pinned major's release notes, and Step 13 judges licences by what ships, not by graph membership. #772
|
|
66
|
+
- **The dead-code and security audits catch a guard whose env var is never set where the guard runs.** `bkit-audit-dead-code` Step 5 — Env Vars now also compares each declared var that gates a control with the secret names set in every runtime environment (the host's env list, the secret-sync map), by name only; a gating var missing from one is a finding, and an environment the token cannot read is `Not Checked`. `bkit-audit-security` Step 14 — Secrets checks that every var a guard reads is in the sync map for each environment that runs it. Dead-code Step 2 reads every unused-import hit in shipped source, Step 3 opens duplicate-export aliases with zero readers and runs once with entry exports included, and Step 9 includes URL pointers outside code. Security Step 6 treats a limiter keyed on a client-chosen field as no limiter, and Step 9 limits a public read of versioned content to published content. #773
|
|
67
|
+
- **The prompts audit checks prompts against `.bkit/authoring.md`.** `bkit-audit-prompts` lists bkit's authoring rules among its anchor standards, read at the installed bkit version, and Step 5 — Structure gives each prompt one row per principle. An install without `.bkit/authoring.md` marks those rows `Not Checked`, never a pass. #774
|
|
68
|
+
- **The architecture audit counts runtime imports only, names dynamic imports as a blind spot, and reads regions.** Step 6 computes client reach over runtime edges, because a type-only import is erased at compile time. Step 7 adds non-literal dynamic import specifiers to the blind spots. Step 9 lists data ownership per script or CLI directory too. Step 10 annotates the Container view with each compute and data region, and a request that crosses regions on every call is a finding. #775
|
|
69
|
+
- **The coverage audit has a fallback when CI keeps no per-file artifact, and samples the integration branch when the default branch has no full runs.** #776
|
|
70
|
+
- **The database audit diffs a schema fingerprint across environments.** Step 5 compares counts and hashes of columns, indexes, functions, policies and grants, because a change made by hand outside migrations leaves both migration lists equal. With one readable environment, the diff is `Not Checked`. #777
|
|
71
|
+
- **The documentation audit catches more broken references.** Step 2 adds a column for cited paths that no longer resolve, Step 4 resolves every `*.md` name (backticked or not) and confirms the prose-lint and format globs cover every live doc, Step 5 flags ledger rows whose `references` is null, and Step 10 runs the path resolver over memory files. #778
|
|
72
|
+
- **The duplication audit searches generic calls and renamed imports, and leaves generated files out of the clone %.** #779
|
|
73
|
+
- **The LLM audit counts effective retries.** Step 11 multiplies wrapper retries, the SDK's default retries and tool-loop steps, because the layers stack, and uses that figure in the worst-case call count. #781
|
|
74
|
+
- **The logs audit stops undercounting the error board and overcounting retries.** Step 3 takes the unresolved count from a per-id lookup and compares the tracker's event retention with the window first. Step 11 names CI as a runtime-log source when CI writes to production resources, and joins retry log lines to the request's final outcome before counting failures. #782
|
|
75
|
+
- **The slop audit flags suppression entries that hide nothing, and searches `void` fire-and-forget calls.** Step 3 prints a hide count per suppression entry, and an entry that hides 0 findings gets the new verdict `PRUNE`. Step 7 searches `void <call>(` in handlers and jobs, because floating-promise rules ignore `void` by default. #783
|
|
76
|
+
|
|
77
|
+
### Fixed
|
|
78
|
+
|
|
79
|
+
- **Workflow summaries judge the findings that remain unresolved.** Later passing work clears only the findings its contract covers; accepted limitations remain concerns, and missing evidence stays unresolved. The original verdicts remain in the assessment history, while current action items reflect the latest result for each task or remediation step.
|
|
80
|
+
- **The slop check ignores skip and focus markers inside literal test data.** Strings, template text and regular expressions containing examples such as `it.skip` no longer count as disabled tests when a valid syntax tree is available; executable markers still count, and an unavailable or invalid syntax tree keeps the conservative text scan. `bkit upgrade` automatically refreshes `scripts/slop/bkit-slop-type-suppression.mjs` only when it is byte-identical to the copy shipped in v6.0.0. Customized copies and symlinked paths stay unchanged, with a visible warning to merge `.bkit/templates/scripts/slop/bkit-slop-type-suppression.mjs` by hand. Dry run reports the update without writing, and a second upgrade changes nothing. See `docs/migrations.md` § Migrations.
|
|
81
|
+
- **Ship reviews removed assertions against their actual surviving coverage.** `bkit branches scan-diff --disabled-tests` recognizes moved assertions once per added occurrence, requires unchanged test context for moves across files, pairs rewrites only within the same file and test, and attributes removals to their original test. Unrelated additions cannot hide a removal. The deferred-work sweep requires concrete evidence of enabled equivalent coverage or an owner-approved retirement that leaves no requirement uncovered; missing or inconclusive coverage stays blocking.
|
|
82
|
+
|
|
83
|
+
- **`/bkit-ship` fixes a CI red that only `bkit:generate` output caused, and pre-push catches a stale register before a PR's CI does** ([#800](https://github.com/cbartens/bkit/issues/800)). When `dev` had not moved, ship skipped the merge and with it `bkit:generate`, so a branch carrying a stale `docs/capabilities.md` reached the merge gate red, and the gate sent every red to the Failure Path. When `dev` has not moved, ship still runs no `bkit:generate` up front and lets CI find a stale generated file. On a red run, `bkit runs gate` (exit 12) now lists each failed step as `{job}/{step}` and reports `regenerable` when every one is a step that `bkit:generate` output fails. Step 10 then runs one regenerate round per merge attempt, without asking, and lets CI re-check. The capability register no longer writes the commit it was built from into `docs/capabilities.md` and the docs that embed its diagrams, because every commit moved that id, so a rebuild changed the file even when nothing else had. The pre-push note on a branch with no open PR now counts source files changed since the last commit that changed the register. Every place that commits `bkit:generate` output (the merge in ship Step 4 and build Step 3, and the regenerate round) now commits only the paths it changed, so another agent's uncommitted work no longer stops the merge from regenerating. The list is a fixed floor (`bkit-tests-static/Check Capability Register`, `bkit-tests-static/Run bkit Check`) plus the new `bkit.testing.ciGenerated` key, where a repo adds its own steps, such as bkit's `bkit-tests-static/Assert Dist Freshness`. Pre-push now runs the register check on a push of a branch with an open PR into `dev`, not only on a push to `dev`, and warns when `gh` is missing or its `pr list` call fails, because it then cannot tell whether a PR is open and never reads that as "no PR". **Re-copy** `scripts/graphify/bkit-graph-pre-push.sh`, `bkit-graphify.sh` and `bkit-graphify-rollup.py` from `.bkit/templates/`, and the stdin loop and `Capability register` call from `.bkit/templates/.husky/pre-push`, because bkit never overwrites your copies. An older hook copy still works, checking the checked-out branch at `HEAD`.
|
|
84
|
+
- **`/bkit-ship` can integrate `dev` into a PR whose branch fell behind** ([#799](https://github.com/cbartens/bkit/issues/799)). Step 4 used to ask GitHub whether the PR was mergeable and then either refresh the head on the server or rebase the branch. Both failed on a normal run: GitHub ignores `merge=union`, so a PR whose ledgers both sides appended showed as CONFLICTING while a local merge was clean, and the rebase was refused because the run's own `status.jsonl` was still uncommitted. Step 4 now runs `bkit branches sync --merge`, which commits the run's uncommitted work, merges `origin/dev` into the branch locally, unions the ledgers, takes a conflicted lockfile from `dev` and regenerates it, runs a declared `bkit:generate`, and leaves only a conflict that needs judgement for `@bkit-ship-conflict-resolve`, which hand-resolves it and re-runs the block. Both merge modes of `bkit branches sync` (`--merge-docs` for build, a `--require-ci` push and land's PR retry; `--merge` for ship) run one driver that differs only in which conflicts it may settle, so a docs merge is committed as `chore({FID}): merge dev` instead of the default merge subject. `bkit branches rebase --hand-resolve` is gone: nothing called it once ship stopped rebasing. The merged branch is pushed plain, so ship never force-pushes, never reads GitHub's `mergeable`, and never requests a server-side update-branch. Step 9 re-validates CI at the merged head as before. `bkit workflows fail` aborts a merge left in progress as it does a rebase.
|
|
85
|
+
- **BREAKING — `bkit ci prose` fails when Vale or its styles are missing, instead of passing without checking anything** ([#765](https://github.com/cbartens/bkit/issues/765)). With no `vale` on `PATH`, or with `.vale/styles/write-good` absent and impossible to restore, the prose gate used to exit 0 with `passed: true` while linting nothing, so CI could go green without checking prose. Both cases now exit nonzero with `passed: false`, and `skipped` still names the cause (`vale-missing` / `write-good-missing`). Linting with no matching files still passes. Every pre-push gate that runs `lint:prose` now fails on a machine without Vale: install it (`brew install vale`, then `vale sync`). The CI and cloud templates install it already. `bkit check` raises the new `vale-not-installed` advisory when `lint:prose` runs `bkit ci prose` and Vale is not on `PATH`.
|
|
86
|
+
- **A table row in `.bkit/workflows.md` no longer loses its end on GitHub.** The debug Step 6 row had a bare `|` in its text, which GitHub reads as a column break, so the end of the row was dropped. The tables in `.bkit/workflows.md`, `.bkit/templates.md`, `.bkit/pipeline.md` and `.bkit/comparison.md` now escape every pipe inside a cell. #741
|
|
87
|
+
- **The test-first audit stops raising test changes and missing reds that are not real.** `bkit tests green` now finds each planned test by name in its file and flags only changes to that test's own lines or to code it runs (a helper it calls, a hook in its `describe`, an import it uses); an edit to another test in the same file is reported as `outside the test only` and is no gap. When the test is not declared exactly once, the whole file counts as before. `bkit tests red` runs a new optional `bkit:lint:file` script on the test before running it and records no red while lint fails, so a lint rewrite never lands after the red; install, `bkit check` and `bkit upgrade` advise `lint-file-script-missing` when `bkit:test:file` is declared without it. A red refused because the test already passes is now written down with HEAD and the uncommitted files; when another task's committed code, sharing a `CREATE` / `UPDATE` file, already made the test pass, `tests green` lists it under `Already landed` instead of raising `no-red`. Plan review and the build challenge reviewers now check by reading that each new test can fail on the code it targets. #723
|
|
88
|
+
|
|
89
|
+
- **GateGuard, ReadGuard and Antigravity's one-time block work for every user on a shared machine.** Their state folders in the system temp folder now carry the user id (`bkit-gateguard-<uid>`, `bkit-readguard-<uid>`, `bkit-antigravity-retry-<uid>`), as the rules-read advisory's does. A second user on a shared Linux `/tmp` no longer finds the folder owned by someone else and silently loses every checklist, repeat-read note and Antigravity advisory. When state still cannot be written, GateGuard gives its checklist as advice only (never a block that no retry can get past), opencode remembers warnings and reads in memory, and Antigravity lets the call through instead of denying it with no way to retry. #700
|
|
90
|
+
- **Tests run by the example pre-push hook can no longer write into your repo's git config.** Pushing from a linked worktree, git hands the hook `GIT_DIR`, which beats a child's cwd and `-C`, so a test's `git -C "$tmp" init` or `git config user.email` landed in the real `.git/config` (`core.bare=true`, a fixture's author identity on every later commit). The example `.husky/pre-push` now clears `GIT_DIR` / `GIT_WORK_TREE` / `GIT_INDEX_FILE` first. Your own copy of the hook is yours: add the same `unset` line after `set -e` to get this. #551
|
|
91
|
+
- **A lettered answer to a bkit question reaches the option it names.** Answering `B` (or `B) Reject`, `B — too risky`) to ship's confirm question for an item the build accepted without the owner no longer records it as "accepted by owner" in `docs/decisions.jsonl`: ship runs `--confirm` only on `Confirm` and `--reject` only on `Reject`, and asks again on anything else. A lettered `Try again` / `Stop` in the debug, build and ship loop questions routes like the option name instead of stalling for another round, and a `Decision: A) Accept …` line counts as accepted, so the next challenge round does not raise the finding again and the build summary counts it. #703
|
|
92
|
+
- **A copy entry outside the checkout no longer breaks every bkit commit.** A `bkit.workflows.worktree.copy` entry such as `../shared/.env`, which setup already skips, was also named in the exclude list of every bkit `git add -A`. Git refused that (exit 128) and nothing was staged, so step commits exited 8, the close commit failed and the resume carry commit could not be made. Such entries are now left out of the exclude list as well, and an abort commit logs a failed add. #662
|
|
93
|
+
- **A file only the main checkout's copy list names is never committed.** Setup copies entries by the main checkout's `package.json`, but bkit commits left out only the entries in the worktree's own list. Commits now leave out the entries of both lists and `.bkit/.env`. #662
|
|
94
|
+
- **A tracked `.env` in the copy list no longer stops every resume with exit 9.** A committed placeholder `.env` that setup copies over shows as modified and is never committed, so the resume rebase refused to start over it. The rebase now sets such entries aside and puts them back (`--autostash`) once the carry commit has left nothing else; a refused carry commit still stops it with exit 9. A worktree holding only copied-in entries also no longer logs a failed carry commit. #662
|
|
95
|
+
- **A resume never deletes a log that differs from `dev`'s copy.** An untracked log of the FID that `dev` already holds is still dropped so `dev`'s copy wins; if its content differs, it is first moved to the feature's outbox folder (`bkit-outbox/{FID}/dropped-logs/`), and the log names where. #662
|
|
96
|
+
- **A design / plan close removes its worktree when the only leftover is the copied `.env`.** `bkit branches land` ignores the `bkit.workflows.worktree.copy` entries and `.bkit/.env` when deciding whether the worktree holds uncommitted work; every repo that does not gitignore `.env` kept `.worktrees/{FID}` and `feat/{FID}` after the close. #662
|
|
97
|
+
- **A `.bkit/.env` copy made where symlinks are refused follows a token rotation.** Setup records the copy it makes and replaces it on the next run, so a rotated token reaches the worktree; only a hand-edited file is left as `differs`. A copy that cannot be written is reported as `bkitEnv: failed` with a log line instead of crashing setup. #702
|
|
98
|
+
- **Setup tips work when the main repo is bare.** With a bare repo plus linked worktrees, tips read the bare repo's parent folder and showed nothing or read an unrelated `package.json`; they now read the worktree's own checkout in that layout. #543
|
|
99
|
+
- **The `bkit-env` setup tip no longer tells you to copy the main checkout's `.bkit/.env` from a worktree.** The tip only fires when that file is missing or blank, so the advice pointed at a file that could not help. #543
|
|
100
|
+
- **Two closes landing at once no longer stop "nothing landed" because they fetched dev at the same moment.** When a second landing in the same clone updated `origin/dev` during this landing's fetch, git failed the fetch with `cannot lock ref 'refs/remotes/origin/dev'` and the close stopped. `pushToDev` now waits and fetches again within `LAND_ATTEMPTS`, as it already does when dev moves. #639
|
|
101
|
+
- **Landing and syncing work in a clone whose fetch refspec does not cover `dev`.** Every dev fetch in the landing, sync and ship-close paths uses `+refs/heads/dev:refs/remotes/origin/dev`, so a single-branch clone no longer builds on a stale `origin/dev` and burns every push attempt. #639
|
|
102
|
+
- **A close no longer tells you to publish a commit that is already on `origin/dev`.** When a later close fast-forwarded local dev past this one, `fastForwardDev` reports `localDev: current` instead of `skipped` with `git pull --rebase && git push` advice; the not-fast-forwarded list names only commits `origin/dev` lacks. #639
|
|
103
|
+
- **Ship's close removes the feature worktree before it releases the workflow lock,** so a new run of the same FID cannot take the lock and resume into a worktree that is about to be deleted. #639
|
|
104
|
+
- **A failed `bkit steps done` / `steps finish` commit leaves the index and `status.jsonl` exactly as it found them, and never reports "nothing written" while its row is still on disk.** The rollback restores the index byte-for-byte: an already-staged `status.jsonl` no longer keeps the refused row, a partly staged file keeps its partial staging, and a bracketed new path such as `src/[id].tsx` no longer unstages your `src/i.tsx`. It removes only the attempt's own row, keeping a row another writer appended during the commit. When the attempt's row cannot come off because another row follows it, the command says the row stays and exits 82 (needs a human) instead of 8, so a re-run cannot write it twice. A hook that deletes `status.jsonl` no longer crashes the command with `ENOENT`. #700
|
|
105
|
+
- **The rules-read advisory on opencode, Codex and Antigravity reaches each session once, even when another user owns the shared temp folder.** The state folder is per user (`bkit-rules-advisory-<uid>`); when state still cannot be written, the opencode plugin remembers the advisory for the life of the process instead of repeating it on every tool call. #700
|
|
106
|
+
- **`context-fetch-unbounded` no longer counts a timeout, a log verbosity or a buffer size as an output bound.** A block such as `curl -sS --max-time 30 "$API/events"` that dumps a whole response read as bounded, because `--max-time`, `curl -m`, `--log-level=` and `--buffer-size=` matched the bound list. These now count for nothing; `grep -m` / `rg -m`, the other `--max-*` caps and every shipped context block still read as bounded. A heredoc holding an apostrophe no longer hides the block's `# Customise:` comment from the check either. #697
|
|
107
|
+
- **Ignoring `context-fetch-missing` no longer hides `context-fetch-unbounded`.** The context-fetch setup tip listed both advisory ids, so one `bkit.advisories.ignore` entry silenced both; the tip now owns only `context-fetch-missing`. #697
|
|
108
|
+
- **Every context-fetch artifact keeps its `===== TRUNCATED: … =====` line through the clamp.** Only `runtime` artifacts kept it; in a large `ci` or `e2e` artifact a block's page-cap notice could be trimmed away and a partial read looked complete. #697
|
|
109
|
+
- **A table-rendered fetch reports the bytes its block printed as fetched.** `clampSummary` counted the markdown table as both fetched and kept, so the saving from rendering never showed. #697
|
|
110
|
+
- **The opencode `variant:` clamp reads each model's thinking levels per gateway, not per model.** The same model exposes different levels on different gateways: `deepseek-v4.1-flash` offers `none` on OpenRouter only, and `kimi-k3` is `max`-only on OpenCode Go. A `low`-effort agent on OpenCode Zen or Go renders `variant: low` instead of the `none` OpenCode 2.0 rejects there, and stays `none` on OpenRouter. Levels come from models.dev `reasoning_options` per provider. #205
|
|
111
|
+
- **`bkit install --opencode` names the models the Ollama and OpenCode Go columns actually pin.** The paid-plan note listed `kimi-k3`, which neither column installs, and left out the models they do; it now reads the pins from the shipped `prompts.md` § Models. #205
|
|
112
|
+
- **The Codex session model for `/bkit-build` and `/bkit-ship` is `gpt-6.1-sol`.** The orchestrator table still named `gpt-5.6-terra`, which no Codex agent pins. #205
|
|
113
|
+
- **README no longer credits the OpenCode Zen and Go DeepSeek pin with the 1017 snapshot's 40.9.** Those gateways serve the undated 2026-09-10 snapshot, which LiveBench 2.0 has not measured. #205
|
|
114
|
+
- **§ Model Presets shows what OpenRouter installs and what Option 2 overrides to.** The summary row and per-agent table named `gemini-4-argon` for Option 2 and `kimi-k3` as shipped; they now match § Models and the copy-paste override block (`gpt-6.1-sol`). The OpenRouter IDs line gives bare IDs: pasting the old `openrouter/xiaomi/…` into an override cell rendered `openrouter/openrouter/…`. #714
|
|
115
|
+
- **The capability-register gate runs on pull requests, not only dev pushes, and judges the tree that lands.** `Check Capability Register` runs on every event as `check-at HEAD`: on a pull request that is GitHub's test merge of the PR into its base, the tree a squash merge puts on dev. A register that drifts in a PR — including a PR behind dev whose register and dev's merge into a file no rebuild produces — reds the PR lane and the capstone instead of the post-merge push run, which the ship gate pins to the squash commit and cannot retro-green. A red on a PR names the base: merge it into the branch, run `pnpm graph:refresh`, commit and push. The graphify release installed and the script that judges now come from the same tree, so a `GRAPHIFY_SPEC` bump on dev no longer reds every PR behind it. Copy the `Check Capability Register` step from `.bkit/templates/.github/workflows/bkit-ci.yml` into your adopted `bkit-ci.yml`. #444
|
|
116
|
+
- **A design / plan worktree is cut from `origin/dev` whatever local `dev` holds, and a design / plan branch already carrying unpushed `dev` commits is refused at Step 2 with the one action that unblocks it.** `bkit worktrees setup` cut every new feature branch from local `dev` when it was a fast-forward of `origin/dev`; a design / plan close lands docs and ledgers on `origin/dev` and nothing else, so a branch carrying another session's unpushed code could not land — `bkit branches land` stopped with exit 17 and a list of paths the run never made. `worktreeBase` now pins design / plan to `origin/dev` (`build`, `debug` and `ship` keep local `dev`, still narrated with the unpushed count), a resumed design / plan branch that carries such commits is refused (exit 7) instead of failing after the whole phase, and both the step-2 refusal and the land refusal name the cause — the unpushed `dev` count, its tip, and `git push origin dev` as the fix — rather than only the paths. #639
|
|
117
|
+
- **A `bkit.workflows.worktree.copy` entry that is also gitignored no longer fails every commit in a worktree.** `addAllButCopied` named each copied entry in a `:(exclude)` pathspec; git refuses an `add` whose pathspec names an ignored file, so `bkit steps finish` — and `workflows handover --push`, `workflows close` and the resume cleanup — exited 8 at Step 2 whenever setup had copied a gitignored entry such as `.bkit/.env`. The helper now leaves an ignored copied entry out of the exclude list (`git add -A` skips it anyway). `bkit branches push --commit` stages through the same helper, so its marker commit no longer commits a copied-in secret; `bkit ledgers commit --path` and `commitOwnSet` name a gitignored path in the error instead of leaking git's ignored-paths guard. #700
|
|
118
|
+
- **Each workflow routes on every verdict its step can give.** Design's apply-review and organise steps and plan's apply-review step went to the Failure Path only on `FAIL`, so any other verdict, such as a `PENDING` left by an agent that stopped early, carried on; they now continue only on `PASS` or `CONCERNS`. Ship integrates `origin/dev` with a local merge and resolves any remaining conflicts before the merge gate. Ship's deferred sweep now gets the FID it is told to read. When the owner answers a debug or ship loop question with something other than an option, the step now answers, asks again and records the wait with a `bkit status row` command. A `FAIL` summary is shown once, as the final message at the STOP step, instead of being shown and then routed to the Failure Path; ship reads its summary before the flush removes the worktree. `bkit upgrade` picks it up. #747
|
|
119
|
+
- **The build summary counts a gaps fix that a later verify passed as answered.** A `FAIL` from `@bkit-build-plan-gaps-remediate` that the group's next verify passed counted as an open failure. Build never routes on the gaps fix's own verdict; the verify after it judges the fixes. #747
|
|
120
|
+
- **Agent verdict rules no longer overlap or leave cases out.** Each prompt's `PASS` / `CONCERNS` / `FAIL` (and `HANDOVER`) rules now cover every outcome exactly once, and `CONCERNS` means "no `FAIL` condition holds" throughout. For example, `@bkit-build-plan-tasks` gives `HANDOVER` over `FAIL`; `@bkit-build-plan-gaps-remediate` and `@bkit-debug-failure-remediate` give `FAIL` and change nothing when Context names no `Strategy`; and `@bkit-ship-conflict-resolve` gives `FAIL` when no merge of `origin/dev` is in progress. `bkit upgrade` picks it up. #747
|
|
121
|
+
- **No audit template contradicts itself any more, and each gate box checks one fact.** The deferrals and failures audits no longer allow ledger rows that their rules forbid. The documentation and duplication audits name the copy to delete or the register row to change, and leave that to the next step, as their other steps say. The annotations audit and seven others say this run writes its own `docs/failures.jsonl` rows. The app-store audit covers a build that is in review. The `Tree clean` gates, which a run could never pass because it writes its own report, are replaced by gates it can pass. The shared `Writes`, `Suggested Issues` and `Summary` gates are split into one fact per box, and long steps into one instruction per line. `bkit upgrade` refreshes `docs/templates/`. #747
|
|
122
|
+
- **The plan template's Gate 1 covers its own items and cites each owner section by its full heading.** Its examples no longer name a vendor, and `Manual Verification` no longer carries the owner mark, since it has no owner confirmation box. `bkit upgrade` refreshes `docs/templates/bkit-plan.md`. #747
|
|
123
|
+
- **The shipped docs match the code.** For example: `providers.md` names `GATEWAY_COLUMN` and `render/hosts.ts`, and says Cursor gets `effort` but does not enforce `tools:`; `prompts.md` says each opencode cell holds the bare model ID and install adds the provider prefix; `cloud.md` says `bkit install` re-stamps the cache-bust line and how the pnpm version is chosen; `install.md` says design, plan and build run in the cloud only by dispatch; `testing.md` and `deployments.md` call the merged-PR push run `partial`. Text true only of the bkit repo, and the outdated v1 model board, are gone from the shipped docs. `bkit upgrade` picks it up. #747
|
|
124
|
+
- **The copy-and-own templates name the files that exist.** The context briefs and `.bkit/templates.md` cite `bkit-maestro.md`, `bkit-expo.md` and the rest by their real names; the capabilities rollup accepts the `hide_nodes_why` key its seed asks for; the bug form says what `[soak]` does without `[repro]`. Re-copy a template you adopted to pick it up. #747
|
|
125
|
+
- **An audit run no longer lowers the severity of an open `docs/failures.jsonl` row.** Step 11 of `@bkit-audit-generate` used to treat any finding at the same `file:line` as the same problem, and a `LOW` finding then rewrote the matched row to `LOW`, so a real high-severity bug that a minor finding only mentioned dropped down every ranking. A finding now matches an open row only when it is the same defect, a matched `MEDIUM`+ finding takes the higher of the two severities, a matched `LOW` finding keeps the row's severity, and no audit run lowers a row's severity. #767
|
|
126
|
+
- **The audit agent follows the `Host shell`, `Shallow history` and `New worktree` rules again.** `@bkit-audit-generate` reads only the project part of `.agents/rules.md`, and gets bkit's own rules through its copied `Rules` block, which had lost these three. They are back, word for word. `Empty is inconclusive` now also says that a capped or windowed read is partial and must be paged to its end, and names two more false empties: `rg` printing only "binary file matches" on history output (use `-a`), and a wildcard `git grep` pathspec that matches the folder but not the files under it (end it with `/**`). `Host shell` now names zsh's `"$var:x"` modifier and its lower-case `pipestatus`, and that a `grep` alias or function can reject options. #768
|
|
127
|
+
|
|
128
|
+
## [6.0.0] — 2026-10-25
|
|
129
|
+
|
|
130
|
+
### Breaking Changes
|
|
131
|
+
|
|
132
|
+
- **BREAKING — `bkit <noun> <verb>` prints text, not JSON; `--json` prints the old JSON.** stdout is `key: value` lines, nested keys indented under their parent, lists of flat rows as markdown tables. A hook, script or CI step that parses bkit output (piped to `jq`, or read with `JSON.parse`) reads nothing until it adds `--json`. The `bkit-cloud-bootstrap.sh` and `bkit-disk-prune.mjs` templates pass `--json`; copy the change into adopted copies. Setup tip `bkit-output-json` names each file still parsing text. Text output also leaves out the keys no workflow reads (each verb's `textOmit`, e.g. the written ledger `row`); `--json` prints every key. #681
|
|
133
|
+
- **BREAKING — the `package.json` `bkit` settings are grouped into sections, and `bkit upgrade` moves them for you.** `tracker` (`default`, `github`, `linear`, `jira`), `cloud.agent`, `providers.opencode` (`default`, `installed`), `testing` (`minPushPaths`, `ciCapstone`, `ciWorkflow`, `shipGate`), `workflows` (`minFreeGb`, `worktree`, `tasks`, and the ask points), `prompts.roadmapNext`, `dependencies.updater` and `advisories.ignore`. On install and upgrade, `migrateBkitLayout` moves each flat key to its section once and deletes the old key; `loopPause` becomes `ask` on every ask point. A value already at the new path wins, and the file keeps its indent. bkit reads only the new layout, so a script of yours that reads an old key from `package.json` reads nothing. `docs/migrations.md` lists every move. #693
|
|
134
|
+
- **BREAKING — `bkit prs merge` refuses a PR whose base is not `dev`.** It exits 8 and merges nothing, and `bkit prs open --base=main` is refused too: bkit opens and merges into `dev` only, and releases stay manual. A consumer script that merges a PR into another base through `bkit prs merge` breaks. #650
|
|
135
|
+
- **BREAKING — `bkit harvest` and `bkit analysis` are removed.** `bkit stats` replaces `bkit analysis` for a repo's own report. Collecting stats across repos is no longer part of the package. #683
|
|
136
|
+
- **BREAKING — `.bkit/.env` names the Jira key `BKIT_JIRA_API_KEY`.** `bkit upgrade` renames `BKIT_JIRA_API_TOKEN` lines in place and keeps the value. #675
|
|
137
|
+
|
|
138
|
+
### Added
|
|
139
|
+
|
|
140
|
+
- **`bkit context fetch` writes a JSON list from a context-fetch block as a markdown table.** It runs the same block runner as `bkit context run-block` (which stays). With `--render=table` (the default), a JSON array of flat objects becomes a table when the table parses back to the same values and types; anything else is written as printed, and stderr logs `render: raw (<reason>)`. The output as printed is copied to `{OUTBOX}/{name}.raw.{ext}` first, never committed; the clamp record gains `render`. `/bkit-debug` Step 4 fetches through it. #681
|
|
141
|
+
- **New verbs and flags for the git work workflows did by hand.** Each prints only the fields the workflow reads next. #679
|
|
142
|
+
- `bkit branches push --mode=plain [--set-upstream] [--ahead-of=<ref>]`: a feature-branch push, never forced; exit 21 when on `dev` / `main` or nothing is ahead.
|
|
143
|
+
- `bkit branches push --commit=<msg> [--label=…] --file --workflow`: one marker commit per run (auto-route suffix from the labels), then the push.
|
|
144
|
+
- `bkit branches fast-forward --onto=<ref>`: fetch and fast-forward; exit 22 when the local branch diverged.
|
|
145
|
+
- `bkit branches scan-diff --base=<ref>` with `--paths=<regex>`, `--marker=<text>` or `--code`: exit 0 when the diff has a match, 1 when not.
|
|
146
|
+
- `bkit branches force-build`: see Fixed.
|
|
147
|
+
- `bkit prs open --file --workflow`: a re-run reuses the PR the first run opened.
|
|
148
|
+
- `bkit workflows start --branch` (exit 15 on another branch; ship clears stale sentinels), `bkit workflows close --summary --summary-missing`, `bkit workflows handover --list-stray`.
|
|
149
|
+
- `bkit worktrees setup --issue` (ship prints `branchWarning`), `bkit counters route --rows=<type>:<next|count>`, `bkit status meta --debug-attempt=run`, `bkit runs gate --fetch`.
|
|
150
|
+
- `bkit ledgers commit --path=<file>` commits one file alone; `--stage-tracked-updates` with a commit mode commits tracked modifications too.
|
|
151
|
+
- **Workflows run from the prompt alone, with no issue, design or plan.** A missing issue no longer stops design, plan, build, debug or ship. Step 1 recovers the issue from a `FID` the prompt names or a `feat/{FID}` branch (`bkit fids parse`); with none, the run warns once (line 1 of its final message) and continues, skipping every tracker read, label, comment, board call and kill-switch check. Workflows still never create an issue. A run with no issue mints its `FID` as `{branch-suffix}-{slug}` with every all-digit word dropped, so no tracker can read it as an issue id; when a finished run already used that name, `bkit fids resolve` takes the first free `-b`, `-c`, … suffix (naming the `FID` in the prompt is how you continue a run). A `feat/{FID}` branch gives its suffix verbatim as the `FID` for any lowercase slug (`feat/upgrade-react-19`), and the issue id in it counts only when `.bkit/issues.jsonl` has a feature row for it; a branch suffix that is not a lowercase slug (`feat/Fix_CI`) stops the run and asks you to rename the branch or name an issue. #650
|
|
152
|
+
- **`bkit fids check --issue` and `bkit steps gate --issue`.** Passing the run's issue id (or an empty value for a run with no issue) checks the `FID` against it: with an issue the `FID` must embed that same id, with none any lowercase slug passes. `bkit prs open --issue` puts only `<!-- bkit:fid={FID} -->` in the PR body when there is no issue. #650
|
|
153
|
+
- **`/bkit-debug` on `dev` works in place, on its own paths and its own commits.** Steps 9 and 10 push nothing. Step 11 pushes only the run's commits with `bkit branches push --mode=dev`, then waits for the `dev` CI run at the pushed commit (`bkit runs dispatch --sha`, `bkit runs wait --until=completed`, `bkit runs gate --capstone-only`) before it releases the lock. A red run, a refused push or a replay conflict goes to the Failure Path naming the run; a file someone else changed during the run stops the workflow and asks you. #650
|
|
154
|
+
- **`bkit stats` writes this repo's stats report to `.bkit/stats.md`.** It reads only this repo's `.bkit/stats.jsonl`, `.bkit/friction.jsonl` and `docs/deferrals.jsonl`, and prints the report too. `chat` / `run` rows are not counted. Always exits 0. #683
|
|
155
|
+
- **Plan caps are a `package.json` setting, `bkit.workflows.tasks`.** `maxPerGroup` (default 3) and `maxTestsPerGroup` (default 15) cap each plan group's tasks and `TEST FIRST` tests; `maxTestsPerTask` (default 10) and `maxTestFilesPerTask` (default 3) cap each task. Install and upgrade add any missing key and never change a set one. A value that is not a whole number of 1 or more falls back to its default, and `bkit check` shows advisory `bkit-tasks-invalid`. `bkit tests plan-check` and `bkit challenge merge` read them; `bkit repo info` prints them as `tasks`, and `/bkit-plan` and `/bkit-build` pass them to every agent that writes, reviews, executes or verifies plan tasks; no prompt carries a number of its own. When `.github/ISSUE_TEMPLATE/bkit-change.yml` shows a cap as `(default N)` with N other than the configured value, `bkit check` shows advisory `bkit-tasks-form-stale` naming the number to change. #668
|
|
156
|
+
|
|
157
|
+
- **`bkit workflows summary` writes each phase's summary.** Design, plan, build, debug and ship run it at their Summary step instead of a summary subagent. It reads the run's `status.jsonl` rows, the plan files, `docs/deferrals.jsonl` and the Decision Pause files, writes `{workflow}-summary.md` and prints `verdict`, `failures`, `reasons` and `nextStep`. FAIL: a build plan task left unticked, a build with no plan file (`No plan file`), the last test run FAIL, or a step whose last row is FAIL with no later step answering it. CONCERNS: a step whose last row is CONCERNS, an item accepted at a Decision Pause, or a deferral written this run. The recommended next command comes from a fixed table per workflow and verdict. Ship's summary shows the PR's size from `gh` (`--pr`), and drops the line when `gh` fails. Ship now runs it after `@bkit-ship-update-docs` returns, not beside it. #682
|
|
158
|
+
- **`bkit challenge merge` turns build challenge findings into plan tasks.** It reads one or two challenge reports, each finding a task block under `## Findings Log` with `Severity`, `Class`, `Fixable`, `Location`, `DONE WHEN` and `TEST FIRST`. Findings at the same `path:line` merge into one (the worse severity wins); a location accepted at a Decision Pause is dropped, including a path with no file extension (`Makefile:12`). A finding at the `Location` and severity of an earlier challenge task that is still open reopens that task instead of adding a twin: its `**Round:**` line gains the round (`1, 2`) and it is listed in `reopened` and `fixTasks`. One already done in an earlier round gets a new task. CRITICAL, HIGH and MEDIUM findings are appended to `{FID}-plan.md` (`--plan`) as task blocks numbered after its last task, in new last groups of at most `bkit.workflows.tasks.maxPerGroup` tasks (default 3). More than 4 x `bkit.workflows.tasks.maxPerGroup` findings writes nothing and routes `pause` with `overflow` set, so the owner picks which to fix. Each LOW finding becomes a `docs/deferrals.jsonl` row with `reason` `low-severity`. The plan is checked before it is written: a failing check (exit 23) changes no file and prints the `problems`; Build Step 10 then re-runs the reviewers once with them, as it does for exit 22. It prints `route`: `pause` when a finding says `Fixable: false`, round 2 left tasks or the findings overflow, `fix` when it appended or reopened tasks, else `none`; also `fixTasks` (reopened, then appended), `appended`, `reopened`, `deferred` (rows written), `dropped`, `plan` and `round`. `overflow` counts only new CRITICAL, HIGH and MEDIUM findings. A report with findings but no task blocks exits 22. Running it again with the same reports appends nothing. Build Step 10 runs it. #682
|
|
159
|
+
- **`bkit status resume --workflow=<name>` names where a stopped run picks up.** It reads `status.jsonl` and prints the step after the active `IN_PROGRESS` run's last `DONE` step (`step`, `number`, `title`, `after`, `group`, `active`, and for build `challengeRound` and `challengeDone`). A build stopped after a Decision Pause resumes where that decision's row sends it: each row records `route` (`fix` → Plan Execute, `verify` → Plan Verify, `tests` → Run Tests). Steps 1-2 run again on every resume, so their rows place no resume point; rows of steps the workflow no longer has are ignored; a step whose next step depends on its verdict resumes at itself; a Plan Execute row with verdict `HANDOVER` resumes at Plan Execute for the same group. A build stopped in one of the removed challenge steps resumes at Run Tests. Every workflow's Resume line calls it. Exit 2 for an unknown workflow, 3 for an unreadable log. #682
|
|
160
|
+
|
|
161
|
+
- **`bkit context brief` writes one build group's brief.** `--plan` and `--group=N` (or `--next`, the first group with an open task) write `build-brief-{G}-{n}.md` into the `--dir` logs folder: the plan's title, Context & Goals, Narrative, Success Criteria and Out of Scope; every group and task with its state; the group's open task blocks word for word; the reds `status.jsonl` (`--file`) already holds for them, each marked "test written and seen failing; code may be partly done"; one line per earlier task they depend on; the plan's `Reuse Inventory` rows and `Architecture & Component Changes` entries that name the open tasks' `CREATE` / `UPDATE` / `DELETE` files; and the `--note-file` text under Deviations. A task is done when its box is ticked or a `build-plan-execute-*-log.md` in the folder has its `DONE — TASK N` line. It prints only `artifact`, `bytes`, `group`, `tasks`, `workBytes` and `workCap` (8192), so the orchestrator never reads the brief. Inventory rows and Architecture entries past the 8 KB work cap are left out and counted, and the brief names the plan section to read instead; task blocks over the cap warn. `--next` with nothing open writes nothing and prints `group: null`. #668
|
|
162
|
+
- **`bkit check` lists plans the next build will convert.** Advisory `plan-task-markers-missing` names each `docs/plans/{YYYYMM}/{FID}-plan.md` with no task markers whose FID has not shipped (`archive/` and `logs/` are not read). Nothing to do by hand: the FID's next `/bkit-build` converts it. #668
|
|
163
|
+
- **Workflow closes record how full each subagent's context got, and `bkit stats` reports it.** On Claude Code, `bkit workflows close` adds `ctx` to the `outcome` row: for each agent type, how many subagents ran (`runs`) and the largest context one model call reached (`peak`), read from the closing session's subagent transcripts that started during the run. `.bkit/stats.md` gains `## Agent context peaks` (agent type, subagents, runs, peak, runs over 160,000 tokens), or "No context data." when no run recorded any. Other hosts record nothing: Cursor keeps no token counts, and Codex, opencode and Antigravity give no confirmed way to tell this run's transcripts apart. A run spread over two Claude Code sessions counts only the closing session. A transcript folder that cannot be read is skipped and never fails the close. #668
|
|
164
|
+
- **`bkit status route --step=` prints `parentSha`,** the commit HEAD was at when the step's last row was written. For a `build-group-start` row that is the group's start commit, so `bkit context diff --base=<parentSha>` reviews only that group's commits. #668
|
|
165
|
+
|
|
166
|
+
- **Every bkit Claude Code subagent summarises its context at 300,000 tokens at the latest.** Each `bkit-*` agent (and the `utility-*` release agents) carries `autoCompactWindow: 300000` in its frontmatter; it only lowers the window, so a lower repo or user setting and `CLAUDE_CODE_AUTO_COMPACT_WINDOW` still win (Claude Code v2.1.296+). Codex, opencode, Cursor and Antigravity have no working per-agent setting, so their renderers drop the field; `.bkit/providers.md` § Context compaction lists each host's answer with its source. Your main session's window stays your setting. #668
|
|
167
|
+
|
|
168
|
+
- **`bkit counters progress`** reports how a fix loop is going since the last decision: problems after each round (`5 → 2`), whether the rounds changed source code, the last and next fix strategy, minutes per round, a trend (`closing` / `stuck` / `worse`) and a recommendation. #670
|
|
169
|
+
- **Step 1 of every phase workflow resolves the issue and `FID` in the engine.** The workflow writes the session prompt to `{SESSION_DIR}/prompt.md` (`bkit repo info` now prints `sessionDir`, a new folder per run under the outbox, pruned after 7 days). `bkit fids parse --prompt-file` prints the issue it names (`source`, `issue`, `ref`, `fidNamed`) for GitHub, Linear or Jira. `bkit fids resolve` replaces the prose chain: a `FID` the prompt names, the `feat/{FID}` branch, the feature logged for the issue (on `dev` too), a mint from the title, then a free `-b`, `-c`, … suffix and the format check, in one call. `bkit fids plans` counts build's sub-plans and tells build and ship when there is no plan. `bkit fids design --description-file --prompt-file` gives plan its parent design and audit report paths; a named file that is missing exits 7. #677
|
|
170
|
+
- **`bkit workflows start` prints where a resumed run picks up.** `lastDoneStep` is the run's last finished step, and `openDecision` the build Decision Pause still waiting for an answer, with every field Step 3 of the pause needs. The workflows no longer read `status.jsonl` or decision files by hand to resume. #677
|
|
171
|
+
- **The engine reads and writes the tracker; the tracker MCP is the fallback.** On GitHub (`gh` with `BKIT_GH_TOKEN`), Linear (its GraphQL API with `BKIT_LINEAR_API_KEY`) and Jira (the Rovo MCP endpoint over HTTP with `BKIT_JIRA_API_KEY`), bkit verbs now post step comments (`steps finish --post`, `issues comment --post`, `workflows fail|handover|close|links --post`), write labels, state and the board move (`issues labels --write`), and read issues. Each prints `tracker`; the workflow calls the tracker MCP only when a verb exits 80 or prints `done` = false. A re-run never posts the same step comment twice: `status.jsonl` records each post as an `engine-action` row, and close-time comments are checked against the issue's last 5 comments. #675
|
|
172
|
+
- **New tracker verbs.** `bkit issues fetch` writes the issue's description, comments and facts to the session folder (or the outbox) and prints only `title`, `url`, `state` and `labels`. `bkit issues related` reads each related-search hit and writes the related-issues file. `bkit issues blocked-by` prints a build's open blockers. `bkit workflows input` writes the run's input file. `bkit runs context` prints `ref`, the issue as `/bkit-auto` passes it on. #675
|
|
173
|
+
- **`bkit workflows close --final-return` prints the run's last message.** It holds the summary, the issue link, `## Suggested Replies` and `## bkit Tips`; `--warning`, `--line` and `--suggest` add the run's own lines. `workflows close` and `workflows fail` delete the run's session folder when given `--session-dir`. #675
|
|
174
|
+
- **`bkit tests run` counts failing tests itself and writes the trimmed log.** It empties `{log}.reports/` next to the log and passes it to the gate as `$BKIT_TEST_REPORT_DIR`; `failures` is the failing plus erroring test cases in the JUnit files written there. A FAIL with no reports prints `failures: null` and a hint, never a guessed count. `--artifact --fid --step --attempt` copies the log into the plan folder, trims it and records the trim in the same call (exit 5 when that fails); build Steps 7 and 14 and debug Step 3 use it, and drop `--failures` when the count is unknown. Wire your test runners' JUnit output into that folder (`docs/testing.md` § JUnit reports for bkit). #678
|
|
175
|
+
- **`bkit status route --prompt-file` decides debug's start.** For `--workflow=debug` it prints `debugAttempt`, `debugMode`, `soakEntry`, `repro` and `gates` from the session prompt and the current run (`--fid` and `--month` stand in for `--file`). Debug's Step 1 is one call instead of jq and prose. `[live=off]` now counts only in the session prompt, not in the issue body. #678
|
|
176
|
+
- **Every trimmed file keeps its untouched original in the outbox.** `context clamp`, `context run-block`, `context diff` and `tests run --artifact` copy the original to `{OUTBOX}/{name}.raw.{ext}` before cutting it (never committed). The clamp record gains `raw`, `source` (the command and inputs that produced the file) and `fetchedAt`. `context clamp` takes `--source`. #678
|
|
177
|
+
- **After-check `github-project-token-scope`.** With a GitHub board configured, `bkit check` warns when `BKIT_GH_TOKEN` cannot write it (a classic token without the `project` scope, or a fine-grained token that cannot see the board), and names the fix. Skipped on cloud runtimes, where the engine never moves the board. #675
|
|
178
|
+
- **Every point where a run can ask you has an `ask` setting and a hard limit.** `package.json` `bkit.workflows.<workflow>.<point>` holds both: `design.grill` (`maxTurns`, default 6), `build.verify`, `build.challenge`, `build.tests`, `debug.fix`, `debug.soak` and `ship.merge` (`maxAttempts`, default 2), each `ask: true` by default. A used-up limit always goes to the Failure Path, and a question never resets a count. With `ask: false`, or in a headless GitHub Actions run, the run takes each recommendation: `Fix now` is applied, and an `Accept` is recorded as assumed. Questions with no recommendation are still asked in a session and go to the Failure Path headless. `bkit counters route --point=` and `bkit counters max --point=` read the settings; a bad value falls back to its default with the advisory `bkit-ask-settings-invalid`. #671
|
|
179
|
+
- **`bkit decisions compose`, `apply` and `pending` handle build questions from pause to ship.** `compose` writes the Decision Pause file in one question layout from the orchestrator's items JSON, with the progress block and the loop question first for `--loop=verify|tests`. `apply --by=headless` takes each recommendation and writes `Accept (assumed)`; it exits 4 when questions with no recommendation are left. `/bkit-ship` lists the assumed items with `pending` at its new Confirm Assumed Items step and asks you to confirm each: `apply --by=owner --confirm` writes the real acceptance, `--reject` records the rejection, and `pending --defer` writes one `needs-decision` deferral per item rejected or never confirmed; ship still merges. #671
|
|
180
|
+
- **`.bkit/authoring.md` holds the authoring rules for every agent-read markdown file.** Docs, prompts and workflows share one set: § All Agent-Read Files, § Notation (one job per punctuation mark, the closed list of capitalised words), § Docs, § Prompts and Workflows, § Prompts Only and § Workflows Only. It ships with the other `.bkit/*.md` docs and has a row in `.agents/docs.md`; `.bkit/prompts.md` and `.bkit/workflows.md` keep only the explanations behind the rules. #684
|
|
181
|
+
|
|
182
|
+
### Changed
|
|
183
|
+
|
|
184
|
+
- **Ship, debug and build make no raw `git` / `gh` calls, `jq` pipes or hand copies.** Every one is now a `bkit` verb, and a test in `workflow-fence-syntax.test.ts` fails on any that comes back (the `{BKIT}` bootstrap line is the one named exception). Ship Step 5 now stops when the fetch of the PR branch fails, where it used to carry on with the local copy. #679
|
|
185
|
+
- **Each agent that reads the run's diff gets a fresh diff file.** Build Steps 7, 9 and 10, debug Step 10, ship Step 5 and every Failure Path handover run `bkit context diff --reset-log` into `tmp/bkit-diff-{FID}.md` right before the agent; the Large Diff Guard rule is gone. Past 10000 lines the file names the files it does not show whole, and the security challenger reads each security-relevant one with its own `git diff`. `context diff` prints `omitted` and takes `--reset-log`. #678
|
|
186
|
+
- **`/bkit-plan` runs one review round with two reviewers and one scope check.** Step 5 runs `@bkit-plan-review-1` (tightness and task order), `@bkit-plan-review-2` (regression risk) and `@bkit-plan-scope-check` (template, criteria and verification) in parallel; Step 6 applies all three reports in one pass to `plan-v2-reviewed.md`. The separate scope round and its apply step are gone. review-1 fails a plan only for a task-order cycle no reordering fixes or a plan that contradicts its design. `bkit upgrade` removes the retired agent files from each host. A plan paused mid-review under the old steps: run `/bkit-plan` again. #665
|
|
187
|
+
- **Build Step 10 runs only the security reviewer on a small change.** `bkit challenge gate` fetches `origin dev` first (offline, it measures against the last fetch and logs that), then measures `origin/dev...HEAD` (`docs/plans/**` excluded) on every pass: under 300 lines and 10 files, `@bkit-build-challenge-2` runs alone and challenge-1 is skipped; a larger change runs both. Each pass writes a `challenge-gate` row to `status.jsonl`, the summary's `## Challenge` section says when challenge-1 was skipped, the `outcome` row carries `build.challenge.gate`, and `bkit stats` counts runs by size. #665
|
|
188
|
+
- **`.bkit/prompts.md` § Model Overrides rows naming an agent bkit no longer ships are reported.** `bkit install`, `check` and `upgrade` list them as the `model-override-unknown-agent` advisory; those rows are not applied. #665
|
|
189
|
+
- **The build challenge is one step, and build has 15 steps.** Step 10 runs the two challenge reviewers, then `bkit challenge merge` adds their findings to the plan as tasks, each with a test to write first. Fixable findings go back through Plan Execute and Plan Verify once (round 2 reviews again but never fixes on its own); a finding no agent can fix, or anything left after round 2, goes to the Decision Pause. Accepting a challenge task there ticks it in the plan. The challenge's `status.jsonl` row records the round it ran as, and `bkit status resume` reports it back as `challengeRound`, so a resumed build picks up the right round. The challenge runs at most two rounds: once a round-2 challenge has merged, `bkit status resume` prints `challengeDone: true` and Step 10 goes on to Run Tests, and a round-2 `Fix now` runs the fix without a third review. A build resumed into an open challenge Decision Pause re-runs the size check and the merge, which finds its tasks already in the plan and writes nothing, and asks the same questions again. The old Steps 11-13 (reconcile, remediate, edge-case tests) are gone, and Steps 14-18 are now 11-15: Run Tests 11, Failure Diagnose 12, Failure Remediate 13, Summary 14, Close Workflow & Create PR 15. `MAX_CHALLENGE_REMEDIATE_ATTEMPT` is removed. #682
|
|
190
|
+
- **`@bkit-build-plan-execute` writes `docs/knowledge.jsonl` gotcha rows** once its tasks' tests pass, now that it fixes challenge findings. #682
|
|
191
|
+
- **`/bkit-plan` writes its tasks in groups and checks the plan after every write; it has 8 steps.** `@bkit-plan-generate` writes groups of task blocks in build order, each group at most 3 tasks and 15 `TEST FIRST` tests. A task's `DEPENDS ON` takes task numbers (`1, 2`) or `TASK N` names. Step 3 runs `bkit tests plan-check --existence=refuse` on the draft straight after the agent, and Step 6 runs it again on `plan-v2-reviewed.md`; a refused plan goes back to the same agent once to fix only the lines the report names, and a second refusal fails the run. A draft still holding `PENDING` (the agent hit its turn cap) gets one more `@bkit-plan-generate` that fills only the `PENDING` sections. Fetch Context no longer checks the draft. Summary is Step 7 and Close Workflow Step 8; a plan stopped after the old Complexity Split resumes at Summary. #668
|
|
192
|
+
- **`bkit tests plan-check` refuses more malformed plans.** Group numbers must ascend through the file, and task numbers within a group, because the build runs groups in file order; neither may be 0. `DEPENDS ON`, `DONE WHEN`, `TEST FIRST` and `TEST EXCEPTION` may each appear once per task. A bold `**TASK N:**` line inside another task's block, and a task depending on itself, are refused. A code fence that is never closed is reported by its line ("nothing after it is read"). A plan with Windows line endings (CRLF) reads like any other, and `DEPENDS ON: none (...)` names no task. #668
|
|
193
|
+
- **`bkit tests green` reports a `DELETE` that did not happen.** A plan task's `DELETE` line whose file still exists, or whose name is still found in the tree (`git grep`, untracked files included, the plan folder skipped), is a `delete-still-referenced` gap. A `DELETE` and `CREATE` of the same thing is a move and is not checked; for a file-only `DELETE` with a `CREATE` of the same file name, the new file and paths to it are not counted as references; when the search itself fails, the gap says `references not checked`. Plan Verify reads each hit: a different thing with the same name is no gap, anything else is `Fixable: true`. A `§ heading` must match a heading line exactly, and in a JSON, YAML or TOML file a dotted key must match its full path (flat or nested, one-line JSON included); in code and other files it matches its last segment. `bkit tests green --group=N` exits 2 when the group is missing or has no tasks. #668
|
|
194
|
+
- **`bkit workflows promote` copies `{FID}-plan.md` only.** It no longer copies or archives `{FID}-plan-N.md` sub-plans and prints no `subPlans`; a `{FID}-plan-N.md` an older run left in the bucket stays where it is and nothing reads it. The newest draft never counts `plan-vN-split.md` or `plan-vN-K.md`, so a re-run of an old split plan promotes its new reviewed plan. `bkit branches land` checks only the one canonical doc. #668
|
|
195
|
+
- **`/bkit-build` runs one plan group at a time.** Steps 6-8 loop over the groups of `{FID}-plan.md`: `bkit context brief --next` writes the group's brief and `@bkit-build-plan-execute` reads only that; verify judges the group's own diff (`bkit context diff --base=` the group's start commit) and is the only step that ticks tasks; the gate between groups writes a `build-group-gate` row (it was `build-subplan-gate`). Verify attempts count per group (`bkit.workflows.build.verify.maxAttempts`). Holistic verify (Step 9) and the challenge (Step 10) run once, after the last group. A Decision Pause `Fix now` writes a task block into the paused group when it stays within 3 tasks and 15 tests, else into a new group right after it; Accept writes a `TEST EXCEPTION` field. `@bkit-build-plan-execute`, `@bkit-build-plan-verify` and `@bkit-build-plan-gaps-remediate` now stop at 100 turns. Near its turn cap `@bkit-build-plan-execute` sets the verdict `HANDOVER`; Step 6 then runs the group once more with a fresh brief that leaves out its `DONE` tasks, counted as a handover, not a failed attempt, and the re-run keeps the existing log. The tracker comment shows `HANDOVER`. #668
|
|
196
|
+
- **A build with no plan, or a plan with no task markers, gets one written first.** Step 3 runs `bkit tests plan-check`; when it finds no plan or no task blocks, the new `@bkit-build-plan-convert` agent writes `{FID}-plan.md` in task-block form, and the check runs again. It works from `build-input-{RUN}.md`, which Step 2 writes for every build with the prompt, the issue text and the plan as it arrived, keeping an existing plan's other sections; where the plan and the issue disagree, the plan wins. Only the convert agent reads `build-input-{RUN}.md`: it holds the whole plan again, so execute, verify, install-skills, the challenge agents and diagnose work from the plan or the group's brief instead, and run instructions in the session prompt (a redo, a resume, a constraint) reach each brief's `Deviations` section through `bkit context brief --note-file`. Ship no longer records `planSource`. A build already past its execute step finishes on the bkit version it started on: upgrade between builds, not during one. On a plan with no task markers `bkit tests green` stops with "this plan has no task markers; run /bkit-build again to convert it" instead of reporting zero tasks. #668
|
|
197
|
+
- **Workflow input files are dated, and the newer text wins.** `bkit workflows input` writes the issue description and each human comment as its own section, each comment with the time it was posted, and dates the plan, design and audit report sections by their last commit's author date (uncommitted changes count as newer than any commit). Every agent that reads an input file is told: where two sections disagree, the newer one wins, and older text still counts wherever nothing newer contradicts it. `###### bkit:ignore-earlier` is still the only way to drop text. `/bkit-build` also runs `@bkit-build-plan-convert` when the issue has comments posted after the plan's last commit, so those comments change the plan before any code; `workflows input` prints the count as `newerThanPlan`. #691
|
|
198
|
+
- **The change issue form shows the plan's full task block.** Its Tasks example and `.bkit/trackers.md` § Issue shapes show every field a plan task has: `CREATE` / `READ` / `UPDATE` / `DELETE`, `DEPENDS ON`, `DONE WHEN`, `TEST FIRST` and `TEST EXCEPTION`, with the group and per-task caps. Free text still works: the build converts it. Copy the updated `.github/ISSUE_TEMPLATE/bkit-change.yml` from `.bkit/templates/` if you use the form. #668
|
|
199
|
+
- **A `TEST EXCEPTION` written in an issue is a request the owner confirms.** The convert agent and `@bkit-plan-generate` copy it into the task word for word and never write one of their own. `bkit tests green` now checks who wrote an exception on a task with no `TEST FIRST` too: one last changed by a workflow step commit is an `exception-refused` gap, which verify marks `Fixable: false`, so the build stops at the Decision Pause. Accept replaces the field with your reason and the date, so its plan-only commit always has a change and git credits you. A plain `TEST FIRST: n/a — reason` is not checked. #668
|
|
200
|
+
- **Subagents carry their bkit rules in the prompt and read only what sits below `bkit:end`.** Each `.bkit/core/prompts/*.md` agent has a `**bkit Rules:**` block holding, word for word, the bkit rules it acts on. Its `**Rules:**` line now reads `.agents/rules.md` below `<!-- bkit:end -->` only (`Universal Rules`, skipping `Orchestrator Rules`): an agent that can search finds that line with `grep -n` and reads from there, so it never loads the bkit block. `bkit-build-plan-execute` also carries `Check gotchas first`. A rule every subagent must follow belongs under `## Project Rules` › `### Universal Rules`; `## Project Settings` still reaches subagents. Workflows still read all of `.agents/rules.md`. #666
|
|
201
|
+
- **A sentence in two or more prompts, or two or more workflows, has one master wording.** It is a clause in `.bkit/instructions.md` § Clauses, and `pnpm instructions:sync` keeps every copy in step: it rewrites near copies and old wordings, and the check fails on a prompt or workflow missing its clause. The cut-off was 5 prompts and 3 workflows; audit templates stay at 5. #684
|
|
202
|
+
- **Plan and audit templates hold shape only; the prompts that act on them carry the rules.** `docs/templates/bkit-plan.md` keeps its headings, tables, checkbox labels, owner callouts and the lines the engine parses, and drops every instruction block (37.4 KB to 15.7 KB, so every new plan is that much smaller). The 35 audit templates drop their shared boilerplate; `bkit-audit-generate` carries it once. `bkit-build-plan-execute` ticks every applicable Pre-Flight and gate box and continues without asking. The `Permissions`, `No placeholders` and `Process Improvements` report boxes each say in a few words what they check. Your `Project Instructions` below `bkit:end` still apply: step titles, numbers and the `Permissions` label are unchanged. #667
|
|
203
|
+
- **Nine rules join `.agents/rules.md`:** New file locations, UI design, Dependencies, Tests discriminate, Test-only files, Explicit logging, CLIs over dashboards, Server-side validation, Multi-store writes. They lived only in the plan template, so agents working without bkit now get them too. #667
|
|
204
|
+
- **Six rules are sharper.** R24 allows appending a log line to your own step output file with `>>`; R43 excepts standing audit parents; R61 lets docs-only landings push straight to `dev` through `bkit branches land`, with a PR only when the push is refused; R62 accepts an explicit empty merge body; R94 excepts a workflow's named decision points; R98 resolves `FID` from tracker markers, or from a `feat/{FID}` branch when no tracker is wired. #666
|
|
205
|
+
- **`bkit-utility-update-tracker` never comments on issues,** reads labels through the tracker MCP first, and lists bugs it finds under `## Found Bugs` for the orchestrator to file; `bkit-utility-update-roadmap` does the same and lists every container with its URL. #666
|
|
206
|
+
- **Every planner prompt that writes `TEST FIRST` names the per-task caps,** as do the issue shapes: 1 to `bkit.workflows.tasks.maxTestsPerTask` tests (default 10) across at most `bkit.workflows.tasks.maxTestFilesPerTask` test files (default 3), the caps `bkit tests plan-check` enforces. A test pins these defaults to the engine's. #667
|
|
207
|
+
- **The `**Code reads:**` line names `grep -l` before `grep -n`,** limits whole-file reads to short files or files being rewritten, and allows a re-read after context was summarised. Prompts with neither `Grep` nor `Bash` no longer carry it. The `**Tools:**` header label is now `**Verify tools:**`. #666
|
|
208
|
+
- **Install and upgrade write every `bkit` setting into `package.json`.** Each setting is now visible where you look for it. A missing key gets its default (for example `workflows.build.verify`, `prompts.roadmapNext: 25`, `testing.minPushPaths`, `workflows.worktree.copy`, the tracker progress and review states); a choice not made yet (`tracker.default`, `cloud.agent`, `tracker.github.project`, `providers.opencode.default`, ...) gets `""` or `[]`, which reads exactly like absent, so the tracker and cloud setup nudges still show. A key already set is never touched, nested blocks get their missing keys, and `bkit check` reports a missing one. #670
|
|
209
|
+
- **Fix loops ask you before each further round.** The build verify → fix-gaps loop and the build tests → diagnose → fix loop stop at the Decision Pause after a failed round while their limit has rounds left, instead of retrying on their own. The pause opens with the progress report and asks whether to run another fix round, worded so "yes" takes its recommendation; **Try again** runs exactly one more round, using the next strategy, and the fixer reads your answer. Whether a loop asks, and how many rounds it gets, are its ask point's `ask` and `maxAttempts` settings (#671). #670
|
|
210
|
+
- **`.bkit/stats.md` is committed.** The managed `.gitignore` block no longer ignores it, so commit the file after `bkit stats`. It holds no timestamp and changes only when the ledgers do. #683
|
|
211
|
+
- **`/bkit-ship` of `dev` or `main` prints the engine's refusal and stops.** It releases its lock first. `bkit worktrees setup` also refuses `main` in place for design, plan and build. #650
|
|
212
|
+
- **`api.github.com` is in the egress allowlist.** The engine's GitHub tracker calls need it on runtimes that filter egress. #675
|
|
213
|
+
|
|
214
|
+
### Fixed
|
|
215
|
+
|
|
216
|
+
- **`/bkit-debug` runs the integration gate when no e2e gate is declared.** Debug Step 3 now tries `bkit:gate:static-unit-integration` before `bkit:gate:static-unit`, so it runs the same gate as build; it used to skip integration tests and could report green on a failure build caught.
|
|
217
|
+
- **The `rm -rf` guard covers the home folder itself.** Every host denies `rm -rf` / `rm -fr` on `~`, `$HOME` and `${HOME}` (bare, with `/` or `/*`), as well as `/`. The hooks also catch `-Rf` and `-r -f`. Only `~/*` and `$HOME/*` were denied before.
|
|
218
|
+
- **`bkit check` passes after `bkit uninstall --opencode` or `--auto`.** A targeted uninstall resets the module's `bkit.providers.opencode` or `bkit.cloud.agent` keys to unset (`""`, `[]`) instead of deleting them, so check no longer reports them as missing.
|
|
219
|
+
- **`[force build]` now reaches CI on a pull request.** CI reads markers from the PR title on `pull_request` events, so the empty `[force build]` commit build Step 18 pushed was ignored. The new `bkit branches force-build` adds ` bkit [force build]` to the open PR's title, then pushes the empty marker commit, whose `synchronize` event starts the run that reads the title. It never amends, and a re-run does not push a second marker. Build calls it after `prs open` when the branch changes a `migrations/` or `migrate/` directory. The consumer CI template reads the marker differently from bkit's own `bkit-ci.yml`: the PR title on `pull_request`, else the head commit message, including on `workflow_dispatch`; bkit's own reads the head commit only on pushes to `main` / `dev`. #679
|
|
220
|
+
- **A finished debug run no longer forces the next run into SOAK.** Soak rows from a closed run used to count on the next run of the same `FID`. #678
|
|
221
|
+
- **A SOAK escalation survives a resumed debug run.** An escalation recorded only in `workflow-meta` used to be lost on resume, and the run restarted in NORMAL. #678
|
|
222
|
+
- **`bkit status route` reads a FAIL with no `failures` as `null`.** It used to print 0. #678
|
|
223
|
+
- **`/bkit-debug` no longer stages the whole tree for its CI-trigger commit.** It runs `git commit --allow-empty` only. On a feature branch, the design, plan, build and debug Failure Paths remove only files the run created by mistake, never other agents' changes. #666
|
|
224
|
+
- **A debug or build after a failed ship starts instead of stopping at Step 2.** A failed ship saves its record to `dev` and leaves the feature worktree as it was; the next resume rebase refused to start over those files (exit 9). `bkit worktrees setup` now commits what an earlier run left uncommitted before that rebase (`chore({FID}): carry the failed ship record left uncommitted bkit`, reported as `carriedSha`). An untracked log `dev` already holds is dropped first so `dev`'s copy wins (`carriedDropped`), and the `bkit.workflows.worktree.copy` entries (such as `.env`) are never committed. Exit 9 now means only that this commit was refused. #662
|
|
225
|
+
- **Failure handovers and close commits never commit the `bkit.workflows.worktree.copy` entries either.** `bkit workflows handover --push` on a feature branch and `bkit workflows close --message` used to commit the whole worktree, so a `.env` that setup copied in and the project had not gitignored could land in a pushed commit. Both now leave those entries out. #662
|
|
226
|
+
- **The ship naming warning fires on a run with no issue.** A branch not named for the `FID` now warns when `ISSUE_ID` is empty; an empty id used to match every branch name. #650
|
|
227
|
+
- **Build Step 15.9 and debug Step 11.9 kill-switch checks skip when `ISSUE_ID` is empty,** and ship Step 14.3's label calls skip on a merged run with no issue. #650
|
|
228
|
+
- **The build boundary gate (Step 7) and the debug local test run (Step 3) can pass their own format check.** Both wrote their test log straight into `docs/plans/{YYYYMM}/logs/{FID}/`, where the gate's `prettier --check .` found the half-written log and failed. They now log to the outbox and copy the finished log in. The `.prettierignore` template gains three `docs/plans/*/logs/*/` lines (`run-tests-*.md`, `build-subplan-gate-*.md`, `context-tests-local-*.md`); templates are copy-and-own, so a consumer that already adopted `.prettierignore` adds the three lines by hand. #650
|
|
229
|
+
|
|
230
|
+
### Removed
|
|
231
|
+
|
|
232
|
+
- **The `bkit-plan-complexity-split` subagent and plan Step 7 Complexity Split.** Groups of task blocks replace sub-plans. `bkit upgrade` deletes its host copies. #668
|
|
233
|
+
- **The chat-stats template (`.bkit/templates/scripts/stats/`).** An adopted `scripts/stats/` copy keeps working on its own; delete it if unused. #683
|
|
234
|
+
- **Three plan subagents: `bkit-plan-review-3`, `bkit-plan-scope-check-1` and `bkit-plan-scope-check-2`.** `/bkit-plan` runs `@bkit-plan-review-1`, `@bkit-plan-review-2` and one `@bkit-plan-scope-check` instead. `bkit upgrade` deletes their host copies; a `## Model Overrides` row naming one of them is reported as `model-override-unknown-agent` and can be deleted. #665
|
|
235
|
+
- **The five summary subagents: `bkit-plan-summary`, `bkit-design-summary`, `bkit-build-summary`, `bkit-debug-summary` and `bkit-ship-summary`.** `bkit workflows summary` writes the summary instead. `bkit upgrade` deletes their host copies; a `## Model Overrides` row naming one of them no longer matches an agent and can be deleted. Past `status.jsonl` and `.bkit/stats.jsonl` rows that name them stay as written. #682
|
|
236
|
+
- **Three build challenge subagents: `bkit-build-challenge-reconcile`, `bkit-build-challenge-remediate` and `bkit-build-challenge-tests`.** `bkit challenge merge` folds the findings into the plan, and Plan Execute fixes and tests them. `bkit upgrade` deletes their host copies; a `## Model Overrides` row naming one of them no longer matches an agent and can be deleted. Past `status.jsonl` rows that name them stay as written. #682
|
|
237
|
+
|
|
238
|
+
---
|
|
239
|
+
|
|
9
240
|
## [5.5.1] — 2026-10-22
|
|
10
241
|
|
|
11
242
|
### Breaking Changes
|