@chankov/agent-fleet 0.0.5 → 0.0.6
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/.pi/extensions/agent-fleet-update-check/README.md +4 -4
- package/.pi/extensions/btw/README.md +10 -10
- package/.pi/extensions/btw/index.ts +11 -11
- package/.pi/extensions/chrome-devtools-mcp/README.md +2 -2
- package/.pi/extensions/chrome-devtools-mcp/index.ts +1 -0
- package/.pi/extensions/compact-and-continue/README.md +2 -2
- package/.pi/extensions/pi-voice-stt/README.md +7 -7
- package/.pi/extensions/pi-voice-stt/index.ts +4 -4
- package/.pi/harnesses/agent-hub/README.md +58 -58
- package/.pi/harnesses/agent-hub/backend-policy.js +1 -1
- package/.pi/harnesses/agent-hub/delegate.ts +3 -3
- package/.pi/harnesses/agent-hub/drift-watchdog.js +1 -1
- package/.pi/harnesses/agent-hub/index.ts +248 -248
- package/.pi/harnesses/agent-hub/package.json +1 -1
- package/.pi/harnesses/agent-hub/run-budget.js +1 -1
- package/.pi/harnesses/coms/README.md +2 -2
- package/.pi/harnesses/coms/index.ts +13 -13
- package/.pi/harnesses/coms/package.json +1 -1
- package/.pi/harnesses/damage-control-continue/README.md +9 -9
- package/.pi/harnesses/damage-control-continue/index.ts +11 -11
- package/.pi/harnesses/damage-control-continue/package.json +1 -1
- package/.pi/prompts/{doctor-agent-fleet.md → af-doctor-agent-fleet.md} +1 -1
- package/.pi/prompts/{setup-agent-fleet.md → af-setup-agent-fleet.md} +1 -1
- package/.versions/0.0.6/.claude/commands/build.md +44 -0
- package/.versions/0.0.6/.claude/commands/code-simplify.md +22 -0
- package/.versions/0.0.6/.claude/commands/design-agent.md +14 -0
- package/.versions/0.0.6/.claude/commands/doctor-agent-fleet.md +17 -0
- package/.versions/0.0.6/.claude/commands/orchestrate.md +112 -0
- package/.versions/0.0.6/.claude/commands/plan.md +16 -0
- package/.versions/0.0.6/.claude/commands/prime.md +22 -0
- package/.versions/0.0.6/.claude/commands/review.md +16 -0
- package/.versions/0.0.6/.claude/commands/set-hermes-telegram.md +15 -0
- package/.versions/0.0.6/.claude/commands/setup-agent-fleet.md +25 -0
- package/.versions/0.0.6/.claude/commands/ship.md +72 -0
- package/.versions/0.0.6/.claude/commands/spec.md +15 -0
- package/.versions/0.0.6/.claude/commands/test.md +19 -0
- package/.versions/0.0.6/.claude/commands/webperf.md +32 -0
- package/.versions/0.0.6/.claude/orchestrate-teams.yaml +24 -0
- package/.versions/0.0.6/.opencode/commands/af-build.md +17 -0
- package/.versions/0.0.6/.opencode/commands/af-code-simplify.md +16 -0
- package/.versions/0.0.6/.opencode/commands/af-compound.md +16 -0
- package/.versions/0.0.6/.opencode/commands/af-design-agent.md +15 -0
- package/.versions/0.0.6/.opencode/commands/af-doctor-agent-fleet.md +17 -0
- package/.versions/0.0.6/.opencode/commands/af-orchestrate.md +112 -0
- package/.versions/0.0.6/.opencode/commands/af-plan.md +16 -0
- package/.versions/0.0.6/.opencode/commands/af-prime.md +22 -0
- package/.versions/0.0.6/.opencode/commands/af-review.md +15 -0
- package/.versions/0.0.6/.opencode/commands/af-set-hermes-telegram.md +15 -0
- package/.versions/0.0.6/.opencode/commands/af-setup-agent-fleet.md +15 -0
- package/.versions/0.0.6/.opencode/commands/af-ship.md +16 -0
- package/.versions/0.0.6/.opencode/commands/af-spec.md +16 -0
- package/.versions/0.0.6/.opencode/commands/af-test.md +21 -0
- package/.versions/0.0.6/.opencode/commands/af-webperf.md +32 -0
- package/.versions/0.0.6/.opencode/orchestrate-teams.yaml +24 -0
- package/.versions/0.0.6/.pi/agents/dispatch-policy.yaml +32 -0
- package/.versions/0.0.6/.pi/agents/model-profiles.yaml +20 -0
- package/.versions/0.0.6/.pi/agents/peers.yaml +146 -0
- package/.versions/0.0.6/.pi/agents/teams.yaml +39 -0
- package/.versions/0.0.6/.pi/damage-control-rules.yaml +297 -0
- package/.versions/0.0.6/.pi/extensions/agent-fleet-update-check/README.md +58 -0
- package/.versions/0.0.6/.pi/extensions/agent-fleet-update-check/index.ts +161 -0
- package/.versions/0.0.6/.pi/extensions/agent-fleet-update-check/package.json +6 -0
- package/.versions/0.0.6/.pi/extensions/btw/README.md +118 -0
- package/.versions/0.0.6/.pi/extensions/btw/index.ts +917 -0
- package/.versions/0.0.6/.pi/extensions/btw/package.json +6 -0
- package/.versions/0.0.6/.pi/extensions/btw/state.js +179 -0
- package/.versions/0.0.6/.pi/extensions/chrome-devtools-mcp/README.md +58 -0
- package/.versions/0.0.6/.pi/extensions/chrome-devtools-mcp/index.ts +93 -0
- package/.versions/0.0.6/.pi/extensions/chrome-devtools-mcp/package.json +6 -0
- package/.versions/0.0.6/.pi/extensions/compact-and-continue/README.md +42 -0
- package/.versions/0.0.6/.pi/extensions/compact-and-continue/index.ts +120 -0
- package/.versions/0.0.6/.pi/extensions/compact-and-continue/package.json +6 -0
- package/.versions/0.0.6/.pi/extensions/mcp-bridge/README.md +46 -0
- package/.versions/0.0.6/.pi/extensions/mcp-bridge/index.ts +206 -0
- package/.versions/0.0.6/.pi/extensions/mcp-bridge/package.json +6 -0
- package/.versions/0.0.6/.pi/extensions/package-lock.json +1143 -0
- package/.versions/0.0.6/.pi/extensions/package.json +9 -0
- package/.versions/0.0.6/.pi/extensions/pi-voice-stt/README.md +201 -0
- package/.versions/0.0.6/.pi/extensions/pi-voice-stt/config.ts +198 -0
- package/.versions/0.0.6/.pi/extensions/pi-voice-stt/examples/stt.json +35 -0
- package/.versions/0.0.6/.pi/extensions/pi-voice-stt/index.ts +248 -0
- package/.versions/0.0.6/.pi/extensions/pi-voice-stt/package.json +6 -0
- package/.versions/0.0.6/.pi/extensions/pi-voice-stt/providers.ts +214 -0
- package/.versions/0.0.6/.pi/extensions/pi-voice-stt/recorder.ts +158 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/README.md +798 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/access-approval.ts +135 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/artifacts.js +92 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/backend-policy.js +153 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/delegate.ts +386 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/drift-watchdog.js +165 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/evidence-rules.js +96 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/footer.ts +29 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/helpers.ts +214 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/index.ts +8045 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/monitor-control.ts +9 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/monitor-lifecycle.ts +13 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/monitor-publisher.ts +59 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/monitor-recovery.ts +3 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/monitor-runtime.ts +3 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/monitor-session-bridge.ts +11 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/package.json +7 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/research-retention.js +34 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/research-watchdog.ts +40 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/return-contract.js +215 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/run-budget.js +155 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/safety-routing.ts +22 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/scope-gate.js +103 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/spawn.ts +246 -0
- package/.versions/0.0.6/.pi/harnesses/agent-hub/version.ts +18 -0
- package/.versions/0.0.6/.pi/harnesses/ask-user-remote/README.md +19 -0
- package/.versions/0.0.6/.pi/harnesses/ask-user-remote/index.ts +321 -0
- package/.versions/0.0.6/.pi/harnesses/ask-user-remote/race-core.js +106 -0
- package/.versions/0.0.6/.pi/harnesses/coms/README.md +92 -0
- package/.versions/0.0.6/.pi/harnesses/coms/index.ts +1745 -0
- package/.versions/0.0.6/.pi/harnesses/coms/package.json +7 -0
- package/.versions/0.0.6/.pi/harnesses/coms/version.ts +18 -0
- package/.versions/0.0.6/.pi/harnesses/damage-control-continue/README.md +117 -0
- package/.versions/0.0.6/.pi/harnesses/damage-control-continue/index.ts +536 -0
- package/.versions/0.0.6/.pi/harnesses/damage-control-continue/package.json +7 -0
- package/.versions/0.0.6/.pi/harnesses/damage-control-continue/version.ts +18 -0
- package/.versions/0.0.6/.pi/harnesses/lib/damage-control-shared.ts +197 -0
- package/.versions/0.0.6/.pi/harnesses/lib/herdr-client.ts +458 -0
- package/.versions/0.0.6/.pi/harnesses/lib/herdr-presence.ts +288 -0
- package/.versions/0.0.6/.pi/harnesses/lib/hermes-monitor-herdr.ts +28 -0
- package/.versions/0.0.6/.pi/harnesses/lib/hermes-monitor-model.ts +9 -0
- package/.versions/0.0.6/.pi/harnesses/lib/hermes-monitor-registry.ts +6 -0
- package/.versions/0.0.6/.pi/harnesses/lib/hermes-monitor-socket.ts +118 -0
- package/.versions/0.0.6/.pi/harnesses/lib/hermes-monitor-store.ts +31 -0
- package/.versions/0.0.6/.pi/harnesses/package-lock.json +35 -0
- package/.versions/0.0.6/.pi/harnesses/package.json +9 -0
- package/.versions/0.0.6/.pi/prompts/af-build.md +24 -0
- package/.versions/0.0.6/.pi/prompts/af-code-simplify.md +22 -0
- package/.versions/0.0.6/.pi/prompts/af-doctor-agent-fleet.md +15 -0
- package/.versions/0.0.6/.pi/prompts/af-plan.md +16 -0
- package/.versions/0.0.6/.pi/prompts/af-review.md +16 -0
- package/.versions/0.0.6/.pi/prompts/af-set-hermes-telegram.md +15 -0
- package/.versions/0.0.6/.pi/prompts/af-setup-agent-fleet.md +27 -0
- package/.versions/0.0.6/.pi/prompts/af-ship.md +17 -0
- package/.versions/0.0.6/.pi/prompts/af-spec.md +15 -0
- package/.versions/0.0.6/.pi/prompts/af-test.md +19 -0
- package/.versions/0.0.6/.pi/skills/bowser/SKILL.md +145 -0
- package/.versions/0.0.6/.version +1 -0
- package/.versions/0.0.6/agents/architect.md +23 -0
- package/.versions/0.0.6/agents/bowser.md +26 -0
- package/.versions/0.0.6/agents/builder.md +53 -0
- package/.versions/0.0.6/agents/code-reviewer.md +205 -0
- package/.versions/0.0.6/agents/deep-researcher.md +31 -0
- package/.versions/0.0.6/agents/documenter.md +18 -0
- package/.versions/0.0.6/agents/orchestrator.md +54 -0
- package/.versions/0.0.6/agents/plan-reviewer.md +60 -0
- package/.versions/0.0.6/agents/planner.md +69 -0
- package/.versions/0.0.6/agents/releaser.md +24 -0
- package/.versions/0.0.6/agents/researcher.md +28 -0
- package/.versions/0.0.6/agents/security-auditor.md +166 -0
- package/.versions/0.0.6/agents/test-engineer.md +151 -0
- package/.versions/0.0.6/agents/web-debugger.md +42 -0
- package/.versions/0.0.6/agents/web-performance-auditor.md +192 -0
- package/.versions/0.0.6/codex/CONDUCTOR.md +148 -0
- package/.versions/0.0.6/docs/ARCHITECTURE.md +210 -0
- package/.versions/0.0.6/docs/agent-fleet-setup.md +366 -0
- package/.versions/0.0.6/docs/claude-code-coms-bridge.md +103 -0
- package/.versions/0.0.6/docs/codex-remote-conductor.md +170 -0
- package/.versions/0.0.6/docs/coms-hermes-bridge.md +179 -0
- package/.versions/0.0.6/docs/getting-started.md +183 -0
- package/.versions/0.0.6/docs/npm-install.md +337 -0
- package/.versions/0.0.6/docs/opencode-setup.md +249 -0
- package/.versions/0.0.6/docs/pi-extensions.md +293 -0
- package/.versions/0.0.6/docs/pi-setup.md +476 -0
- package/.versions/0.0.6/docs/skill-anatomy.md +150 -0
- package/.versions/0.0.6/hermes/README.md +231 -0
- package/.versions/0.0.6/hermes/desktop-plugins/agent-fleet-monitor/plugin.js +19 -0
- package/.versions/0.0.6/hermes/desktop-plugins/agent-fleet-monitor/state.js +10 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter-cap.test.py +15 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter-discovery.test.py +19 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter-envelope.test.py +13 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter-forged-owner.test.py +11 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter-lease-list.test.py +9 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter-multihub.test.py +17 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter-output.test.py +34 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter.py +151 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/adapter.test.py +39 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/manifest.json +11 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/plugin-api-validation.test.py +12 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/plugin_api.py +50 -0
- package/.versions/0.0.6/hermes/plugins/agent-fleet-monitor/dashboard/review-live-wiring.red.test.py +22 -0
- package/.versions/0.0.6/hermes/skills/hub-conductor/SKILL.md +84 -0
- package/.versions/0.0.6/hermes/skills/hub-liaison/SKILL.md +84 -0
- package/.versions/0.0.6/hooks/SIMPLIFY-IGNORE.md +90 -0
- package/.versions/0.0.6/hooks/coms-stop-hook.mjs +73 -0
- package/.versions/0.0.6/hooks/hooks.json +14 -0
- package/.versions/0.0.6/hooks/session-start.sh +74 -0
- package/.versions/0.0.6/hooks/simplify-ignore.sh +302 -0
- package/.versions/0.0.6/justfile +245 -0
- package/.versions/0.0.6/references/accessibility-checklist.md +160 -0
- package/.versions/0.0.6/references/definition-of-done.md +67 -0
- package/.versions/0.0.6/references/fleet-coordination-patterns.md +151 -0
- package/.versions/0.0.6/references/observability-checklist.md +91 -0
- package/.versions/0.0.6/references/orchestration-patterns.md +370 -0
- package/.versions/0.0.6/references/performance-checklist.md +153 -0
- package/.versions/0.0.6/references/prompting-patterns.md +380 -0
- package/.versions/0.0.6/references/security-checklist.md +179 -0
- package/.versions/0.0.6/references/testing-patterns.md +236 -0
- package/.versions/0.0.6/scripts/codex-conductor.ts +41 -0
- package/.versions/0.0.6/scripts/codex-remote-control.ts +167 -0
- package/.versions/0.0.6/scripts/coms-claude-bridge.ts +286 -0
- package/.versions/0.0.6/scripts/coms-cli.ts +484 -0
- package/.versions/0.0.6/scripts/coms-hermes-bridge.ts +413 -0
- package/.versions/0.0.6/scripts/fleet.ts +141 -0
- package/.versions/0.0.6/scripts/hermes-monitor-capability-runner.ts +25 -0
- package/.versions/0.0.6/scripts/hermes-monitor-scenario.ts +7 -0
- package/.versions/0.0.6/scripts/lib/claude-bridge-core.ts +100 -0
- package/.versions/0.0.6/scripts/lib/codex-conductor.ts +88 -0
- package/.versions/0.0.6/scripts/lib/codex-remote-control.ts +480 -0
- package/.versions/0.0.6/scripts/lib/coms-envelope.ts +539 -0
- package/.versions/0.0.6/scripts/lib/fake-herdr-server.ts +88 -0
- package/.versions/0.0.6/scripts/lib/fleet-command.ts +109 -0
- package/.versions/0.0.6/scripts/lib/herdr-layout.ts +263 -0
- package/.versions/0.0.6/scripts/lib/hermes-bridge-core.ts +241 -0
- package/.versions/0.0.6/scripts/lib/hermes-monitor-herdr.ts +2 -0
- package/.versions/0.0.6/scripts/lib/hermes-monitor-model.ts +3 -0
- package/.versions/0.0.6/scripts/lib/hermes-monitor-registry.ts +2 -0
- package/.versions/0.0.6/scripts/lib/hermes-monitor-socket.ts +2 -0
- package/.versions/0.0.6/scripts/lib/hermes-monitor-store.ts +2 -0
- package/.versions/0.0.6/scripts/lib/persona-banner.ts +62 -0
- package/.versions/0.0.6/scripts/lib/spawn-stagger.ts +101 -0
- package/.versions/0.0.6/scripts/lib/team-project.ts +191 -0
- package/.versions/0.0.6/scripts/lib/team-snapshot.ts +147 -0
- package/.versions/0.0.6/scripts/peer-banner.ts +40 -0
- package/.versions/0.0.6/scripts/team-snapshot.ts +237 -0
- package/.versions/0.0.6/scripts/team-up.ts +351 -0
- package/.versions/0.0.6/skills/_internal/grilling.md +11 -0
- package/.versions/0.0.6/skills/browser-testing-with-devtools/SKILL.md +352 -0
- package/.versions/0.0.6/skills/code-review-and-quality/SKILL.md +381 -0
- package/.versions/0.0.6/skills/compound-learning/SKILL.md +137 -0
- package/.versions/0.0.6/skills/context-engineering/SKILL.md +291 -0
- package/.versions/0.0.6/skills/deprecation-and-migration/SKILL.md +206 -0
- package/.versions/0.0.6/skills/designing-agents/SKILL.md +394 -0
- package/.versions/0.0.6/skills/designing-agents/pi-harness-authoring.md +222 -0
- package/.versions/0.0.6/skills/frontend-ui-engineering/SKILL.md +328 -0
- package/.versions/0.0.6/skills/git-workflow-and-versioning/SKILL.md +316 -0
- package/.versions/0.0.6/skills/guided-workspace-setup/SKILL.md +544 -0
- package/.versions/0.0.6/skills/guided-workspace-setup/companion-manifest.json +35 -0
- package/.versions/0.0.6/skills/idea-refine/SKILL.md +182 -0
- package/.versions/0.0.6/skills/idea-refine/examples.md +238 -0
- package/.versions/0.0.6/skills/idea-refine/frameworks.md +99 -0
- package/.versions/0.0.6/skills/idea-refine/refinement-criteria.md +113 -0
- package/.versions/0.0.6/skills/idea-refine/scripts/idea-refine.sh +15 -0
- package/.versions/0.0.6/skills/incremental-implementation/SKILL.md +287 -0
- package/.versions/0.0.6/skills/interview-me/SKILL.md +226 -0
- package/.versions/0.0.6/skills/orchestration-verification/SKILL.md +181 -0
- package/.versions/0.0.6/skills/peer-coms/SKILL.md +90 -0
- package/.versions/0.0.6/skills/performance-optimization/SKILL.md +350 -0
- package/.versions/0.0.6/skills/planning-and-task-breakdown/SKILL.md +236 -0
- package/.versions/0.0.6/skills/security-and-hardening/SKILL.md +461 -0
- package/.versions/0.0.6/skills/spec-driven-development/SKILL.md +245 -0
- package/.versions/0.0.6/skills/using-agent-skills/SKILL.md +199 -0
- package/.versions/0.0.6/systemd/user/agent-fleet-codex-remote-control.service.in +23 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/LICENSE +21 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/README.md +11 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/api-and-interface-design/SKILL.md +294 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/browser-testing-with-devtools/SKILL.md +317 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/ci-cd-and-automation/SKILL.md +390 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/code-review-and-quality/SKILL.md +396 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/code-simplification/SKILL.md +331 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/context-engineering/SKILL.md +289 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/debugging-and-error-recovery/SKILL.md +300 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/deprecation-and-migration/SKILL.md +247 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/documentation-and-adrs/SKILL.md +278 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/doubt-driven-development/SKILL.md +243 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/frontend-ui-engineering/SKILL.md +328 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/git-workflow-and-versioning/SKILL.md +355 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/idea-refine/SKILL.md +178 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/idea-refine/examples.md +238 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/idea-refine/frameworks.md +99 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/idea-refine/refinement-criteria.md +113 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/idea-refine/scripts/idea-refine.sh +15 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/incremental-implementation/SKILL.md +249 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/interview-me/SKILL.md +225 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/observability-and-instrumentation/SKILL.md +203 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/performance-optimization/SKILL.md +350 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/planning-and-task-breakdown/SKILL.md +234 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/security-and-hardening/SKILL.md +467 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/shipping-and-launch/SKILL.md +310 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/source-driven-development/SKILL.md +194 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/spec-driven-development/SKILL.md +206 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/test-driven-development/SKILL.md +383 -0
- package/.versions/0.0.6/vendor/agent-skills-upstream/skills/using-agent-skills/SKILL.md +191 -0
- package/CHANGELOG.md +6 -0
- package/README.md +24 -23
- package/agents/documenter.md +1 -1
- package/agents/releaser.md +1 -1
- package/bin/cli.js +20 -20
- package/bin/lib/bootstrap.js +16 -16
- package/bin/lib/doctor.js +1 -1
- package/docs/ARCHITECTURE.md +4 -4
- package/docs/agent-fleet-setup.md +11 -11
- package/docs/coms-hermes-bridge.md +8 -8
- package/docs/npm-install.md +21 -21
- package/docs/pi-extensions.md +7 -7
- package/docs/pi-setup.md +51 -47
- package/hermes/README.md +1 -1
- package/package.json +1 -1
- package/references/fleet-coordination-patterns.md +2 -2
- package/skills/browser-testing-with-devtools/SKILL.md +1 -1
- package/skills/compound-learning/SKILL.md +3 -3
- package/skills/guided-workspace-setup/SKILL.md +16 -14
- /package/.pi/prompts/{build.md → af-build.md} +0 -0
- /package/.pi/prompts/{code-simplify.md → af-code-simplify.md} +0 -0
- /package/.pi/prompts/{plan.md → af-plan.md} +0 -0
- /package/.pi/prompts/{review.md → af-review.md} +0 -0
- /package/.pi/prompts/{set-hermes-telegram.md → af-set-hermes-telegram.md} +0 -0
- /package/.pi/prompts/{ship.md → af-ship.md} +0 -0
- /package/.pi/prompts/{spec.md → af-spec.md} +0 -0
- /package/.pi/prompts/{test.md → af-test.md} +0 -0
|
@@ -38,7 +38,7 @@ it here, and the reader picks it up.
|
|
|
38
38
|
built-in default.
|
|
39
39
|
- **Validation.** Because an unknown section or key silently falls back to the
|
|
40
40
|
default, typos are invisible at runtime. `agent-fleet doctor` (and the
|
|
41
|
-
|
|
41
|
+
runtime's Agent Fleet doctor command) validates the file — unknown sections,
|
|
42
42
|
unknown keys in known sections, invalid values for the mechanically parsed
|
|
43
43
|
`agent-hub` keys, missing `rules:` folders, and unset `## env` vars — as
|
|
44
44
|
**advisory, warn-only findings**; it never edits the file.
|
|
@@ -91,7 +91,7 @@ override files keep working unchanged. When both sections are present their keys
|
|
|
91
91
|
with later lines winning.
|
|
92
92
|
|
|
93
93
|
The `rules:` and `docs:` keys are also read outside the harness: the `compound-learning`
|
|
94
|
-
skill (and the `/compound`
|
|
94
|
+
skill (and the `/compound` Claude Code / `/af-compound` OpenCode commands built on it) resolves them as the
|
|
95
95
|
targets an end-of-session compound pass writes lessons to.
|
|
96
96
|
|
|
97
97
|
| Key | Default | Meaning |
|
|
@@ -99,21 +99,21 @@ targets an end-of-session compound pass writes lessons to.
|
|
|
99
99
|
| `language` | `English` | User-facing language the dispatcher uses for every `ask_user` question, every `context` field, and every summary. Specialist task strings always stay in English regardless. |
|
|
100
100
|
| `persona-gate` | `off` | When `on`, blocks input at session start until an orchestrator persona is picked. |
|
|
101
101
|
| `model.<persona>` | persona frontmatter `model:` | Replaces the named persona's default model for this project (a full pi model spec). |
|
|
102
|
-
| `models.<persona>` | persona frontmatter `models:` | Replaces the named persona's model-candidate list for `/agent-model` and `/models` profiles (comma-separated pi model specs). |
|
|
103
|
-
| `thinking.<persona>` | persona frontmatter `thinking:` | Replaces the named persona's pi `--thinking` reasoning level for this project: one of `off`, `minimal`, `low`, `medium`, `high`, `xhigh`. Switchable at runtime with `/agent-model-thinking <persona>`. An invalid value is ignored with a session-start warning. |
|
|
102
|
+
| `models.<persona>` | persona frontmatter `models:` | Replaces the named persona's model-candidate list for `/af-agent-model` and `/af-models` profiles (comma-separated pi model specs). |
|
|
103
|
+
| `thinking.<persona>` | persona frontmatter `thinking:` | Replaces the named persona's pi `--thinking` reasoning level for this project: one of `off`, `minimal`, `low`, `medium`, `high`, `xhigh`. Switchable at runtime with `/af-agent-model-thinking <persona>`. An invalid value is ignored with a session-start warning. |
|
|
104
104
|
| `subagents.<persona>.<role>` | persona frontmatter `subagents:` | Replaces or adds one delegate sub-role for this project: `<model>[, tools=<caps>]`. Other declared roles keep their frontmatter values. |
|
|
105
105
|
| `delegate-depth.<persona>` | persona frontmatter `delegate_depth:` (default/max 1) | Replaces the persona's delegation depth budget: `0` makes its delegate tool refuse (delegation off for this project), `1` lets it spawn terminal children. Values above 1 are clamped to 1; children at remaining depth 0 do not receive delegate tooling. |
|
|
106
106
|
| `rules` | none | Comma-separated repo-relative folders holding the project's own rule files (HOW — implementation patterns the work must comply with). Resolution is **index-first**: when a folder has a top-level `README.md`/`index.md`, personas read it first and follow its loading manifest (session bundles, conditional-load lists) instead of bulk-reading the tree; a folder without an index is searched **recursively** through all subfolders. The harness tells every dispatched specialist where the rules live and how to resolve them; the planner and code-reviewer personas read the relevant rules, validate their subject against them, and pass them on (cited in plan acceptance criteria / handed to delegate sub-reviewers). Missing folders produce a session-start warning. |
|
|
107
107
|
| `docs` | none | Comma-separated repo-relative documentation **entry points** (WHAT/WHY — architecture, standards, decisions): canonical files (e.g. `Docs/AGENTS.md`) or doc folders (personas start from a folder's `README.md`/index). Unlike `rules`, docs orient rather than bind: every dispatched specialist and research helper is told to read the entry points relevant to its task and follow their links instead of bulk-reading doc trees; the code-reviewer flags changes that alter documented behavior without a doc update; the documenter treats the entry points and the trees they link as the docs it maintains. Missing paths produce a session-start warning. |
|
|
108
|
-
| `research-keep` | `4` | How many **finished** manual/persona research helpers the hub retains (LRU by finish time) so they stay resumable via `/agents-cont rN`; older ones are pruned along with their session files. `all` disables the cap. Auto-research pipe helpers (spawned for `NEEDS_RESEARCH:` pauses) are always pruned as soon as they finish — their findings persist as files under `.pi/agent-sessions/findings/`. Running helpers are never pruned. An invalid value is ignored with a session-start warning. |
|
|
108
|
+
| `research-keep` | `4` | How many **finished** manual/persona research helpers the hub retains (LRU by finish time) so they stay resumable via `/af-agents-cont rN`; older ones are pruned along with their session files. `all` disables the cap. Auto-research pipe helpers (spawned for `NEEDS_RESEARCH:` pauses) are always pruned as soon as they finish — their findings persist as files under `.pi/agent-sessions/findings/`. Running helpers are never pruned. An invalid value is ignored with a session-start warning. |
|
|
109
109
|
| `recon-search-timeout-s` | `120` | Parent-side deadline, in seconds, for each `read`, `grep`, `find`, or `ls` call made by a native research helper or nested delegate child. Accepts an integer `1`–`3600` or `off`; invalid input warns and falls back to `120`. This is **not** an agent/turn deadline: non-tool work can run indefinitely. On timeout the hub returns `tool_timeout` with call metadata, sends SIGTERM to its owned process group, escalates to SIGKILL after a finite grace period, and settles even if the child never closes or a descendant holds a pipe. If that bounded cleanup cannot confirm child/process-group death, the metadata reports `terminationConfirmed: false`; this prevents a parent hang but may indicate an uninterruptible OS-level process needing operator attention. |
|
|
110
|
-
| `mode` | `standard` | Execution mode for the hub's per-turn budgets: `fast` (single-specialist path, no nested delegation, assertion ledger optional), `standard` (batched execution with moderate budgets), or `strict` (full Verification Contract with wide budgets). Budgets are per **user turn** — exhausted budgets make `dispatch_agent`/`spawn_research` refuse with "summarize and ask the user"; the next user message opens a fresh window. Switchable live with `/hub-mode`. |
|
|
110
|
+
| `mode` | `standard` | Execution mode for the hub's per-turn budgets: `fast` (single-specialist path, no nested delegation, assertion ledger optional), `standard` (batched execution with moderate budgets), or `strict` (full Verification Contract with wide budgets). Budgets are per **user turn** — exhausted budgets make `dispatch_agent`/`spawn_research` refuse with "summarize and ask the user"; the next user message opens a fresh window. Switchable live with `/af-hub-mode`. |
|
|
111
111
|
| `max-dispatches-per-turn` | mode default (2/8/24) | `dispatch_agent` calls allowed per user turn. Positive integer or `off` (unlimited). |
|
|
112
|
-
| `max-research-per-turn` | mode default (1/4/12) | `spawn_research` calls allowed per user turn (the auto-research pipe and the `/research` command are exempt). Positive integer or `off`. |
|
|
112
|
+
| `max-research-per-turn` | mode default (1/4/12) | `spawn_research` calls allowed per user turn (the auto-research pipe and the `/af-research` command are exempt). Positive integer or `off`. |
|
|
113
113
|
| `turn-wall-time-s` | mode default (900/3600/14400) | Wall-clock budget, in seconds, per user turn; once exceeded, further dispatch/research calls refuse. Positive integer or `off`. |
|
|
114
114
|
| `agent-turn-timeout-s` | mode default (600/1800/off) | Whole-run deadline, in seconds, for each spawned specialist, research helper, and nested delegate child (unlike `recon-search-timeout-s`, this bounds the entire run). On expiry the run terminates as `turn_timeout` (exit 124) with partial output preserved. Positive integer or `off`. |
|
|
115
115
|
| `session-recycle-runs` | mode default (3/5/5) | Recycle a specialist's accumulated session (fresh spawn instead of `-c` resume) after this many resumed runs. Context is also always recycled at ≥60% measured context (input + cacheRead + cacheWrite). Positive integer or `off` (context threshold still applies). |
|
|
116
|
-
| `watchdog` | `auto` | Drift watchdog default for dispatched specialists: `auto`/`on` arm the in-flight rules (out-of-scope writes, tool-call loops, repeated failures, tool-call cap) with LLM-judge escalation; `off` disarms. Overridable live per hub (`/watchdog on\|off\|auto`), per agent (`/watchdog <agent> on\|off\|clear`), and per dispatch (the `watchdog` param). A DRIFTING/STUCK verdict terminates the run as `drift_stop` (exit 125) with partial output preserved. |
|
|
116
|
+
| `watchdog` | `auto` | Drift watchdog default for dispatched specialists: `auto`/`on` arm the in-flight rules (out-of-scope writes, tool-call loops, repeated failures, tool-call cap) with LLM-judge escalation; `off` disarms. Overridable live per hub (`/af-watchdog on\|off\|auto`), per agent (`/af-watchdog <agent> on\|off\|clear`), and per dispatch (the `watchdog` param). A DRIFTING/STUCK verdict terminates the run as `drift_stop` (exit 125) with partial output preserved. |
|
|
117
117
|
| `watchdog-judge-model` | researcher persona's model | pi model spec for the one-shot drift judge (e.g. `openai-codex/gpt-5.3-codex-spark`). Falls back to the researcher persona's resolved model, then the dispatcher's. |
|
|
118
118
|
|
|
119
119
|
Example — switch the dispatcher to Bulgarian, pin the builder to sonnet, raise the
|
|
@@ -152,7 +152,7 @@ required: APP_TEST_ADMIN_USER, APP_TEST_ADMIN_PASS, AZURE_SPEECH_KEY
|
|
|
152
152
|
```
|
|
153
153
|
|
|
154
154
|
No skill or harness loads this section. Its only reader is `agent-fleet doctor`
|
|
155
|
-
(and
|
|
155
|
+
(and the runtime's Agent Fleet doctor command), which warns when a declared name is neither set in
|
|
156
156
|
the environment nor declared in the workspace root `.env`.
|
|
157
157
|
|
|
158
158
|
### Per-peer env files (`env_file:` in peers.yaml)
|
|
@@ -243,12 +243,12 @@ available). The reader differs by runtime: pi's harness parses its YAML, while
|
|
|
243
243
|
| pi | `.pi/agents/teams.yaml` | via the `agent-hub` harness (no `/orchestrate`) |
|
|
244
244
|
|
|
245
245
|
Switch teams at runtime with `/orchestrate <team> "<task>"` (parallel of pi's
|
|
246
|
-
`/agents-team`); use `/orchestrate team=<name> <task>` to disambiguate when the
|
|
246
|
+
`/af-agents-team`); use `/orchestrate team=<name> <task>` to disambiguate when the
|
|
247
247
|
task text starts with a word that collides with a team key.
|
|
248
248
|
|
|
249
249
|
**Guided-setup behaviour.** `guided-workspace-setup` offers `/orchestrate`
|
|
250
250
|
**`★`-recommended** for claude-code + opencode (hidden for pi via the existing
|
|
251
|
-
source-availability filter — there is no `.pi/prompts/orchestrate.md`), and
|
|
251
|
+
source-availability filter — there is no `.pi/prompts/af-orchestrate.md`), and
|
|
252
252
|
installs the agent's `orchestrate-teams.yaml` as a **companion** of the command
|
|
253
253
|
(a user-edited config is preserved on uninstall, never silently clobbered).
|
|
254
254
|
Two artifacts stay **claude-only**: the lifecycle **hooks** (they register into
|
|
@@ -101,19 +101,19 @@ just fleet conductor hermes docs --dry-run # preview; no Herdr calls
|
|
|
101
101
|
|
|
102
102
|
The live recipe reuses `scripts/team-up.ts --conductor`: it creates a normal herdr workspace labeled `<worktree-tag>-conductor-<team>` (the tag is the last dot-segment of the checkout's basename, so the same team from a different worktree gets its own workspace), places a `conductor` pane running `hermes -p dev`, and tiles the chosen team beside it. Team peers keep their normal coms harness and herdr presence reporting, so they continue to show agent state in the sidebar; Hermes' own herdr-agent-state plugin is responsible for the conductor pane's state.
|
|
103
103
|
|
|
104
|
-
###
|
|
104
|
+
### Telegram bridge control command
|
|
105
105
|
|
|
106
|
-
The outbound `hermes send` path needs a configured Telegram channel but no skill. Full reply round-trip additionally needs Agent Fleet's `hub-liaison` skill in the profile that owns the running Telegram gateway.
|
|
106
|
+
The outbound `hermes send` path needs a configured Telegram channel but no skill. Full reply round-trip additionally needs Agent Fleet's `hub-liaison` skill in the profile that owns the running Telegram gateway. Pi and OpenCode use the namespaced command:
|
|
107
107
|
|
|
108
108
|
```text
|
|
109
|
-
/set-hermes-telegram status --profile default
|
|
110
|
-
/set-hermes-telegram install --profile default
|
|
111
|
-
/set-hermes-telegram install --profile default --force --restart
|
|
112
|
-
/set-hermes-telegram on 7883056502:1735 --profile default
|
|
113
|
-
/set-hermes-telegram off 7883056502:1735
|
|
109
|
+
/af-set-hermes-telegram status --profile default
|
|
110
|
+
/af-set-hermes-telegram install --profile default
|
|
111
|
+
/af-set-hermes-telegram install --profile default --force --restart
|
|
112
|
+
/af-set-hermes-telegram on 7883056502:1735 --profile default
|
|
113
|
+
/af-set-hermes-telegram off 7883056502:1735
|
|
114
114
|
```
|
|
115
115
|
|
|
116
|
-
|
|
116
|
+
Claude Code exposes the unprefixed `/set-hermes-telegram`; the arguments and backend are identical. The deterministic CLI can also be called directly with `agent-fleet set-hermes-telegram ...`.
|
|
117
117
|
|
|
118
118
|
#### `status` and profile resolution
|
|
119
119
|
|
package/docs/npm-install.md
CHANGED
|
@@ -32,12 +32,13 @@ So:
|
|
|
32
32
|
```bash
|
|
33
33
|
# In the workspace you want to configure:
|
|
34
34
|
npx @chankov/agent-fleet init
|
|
35
|
-
# Then open your coding agent in this directory and run:
|
|
36
|
-
# /setup-agent-fleet
|
|
35
|
+
# Then open your coding agent in this directory and run the command it prints:
|
|
36
|
+
# Claude Code: /setup-agent-fleet
|
|
37
|
+
# Pi/OpenCode: /af-setup-agent-fleet
|
|
37
38
|
```
|
|
38
39
|
|
|
39
40
|
That's it. `npx` fetches the package, the CLI detects your coding agent and
|
|
40
|
-
prints the
|
|
41
|
+
prints the runtime-appropriate command, which runs the full guided install
|
|
41
42
|
inside your agent.
|
|
42
43
|
|
|
43
44
|
### First-class pi package
|
|
@@ -63,27 +64,28 @@ The package also ships assets for the **experimental Linux Codex Remote-Control
|
|
|
63
64
|
### `npx @chankov/agent-fleet init`
|
|
64
65
|
|
|
65
66
|
Materializes the package, **bootstraps the installer artifacts** into the
|
|
66
|
-
workspace
|
|
67
|
-
and
|
|
67
|
+
workspace, and hands off to the runtime's Agent Fleet setup command. Claude Code
|
|
68
|
+
uses unprefixed installer commands; Pi and OpenCode use `/af-setup-agent-fleet`
|
|
69
|
+
and `/af-doctor-agent-fleet`.
|
|
68
70
|
|
|
69
71
|
What `init` writes per agent:
|
|
70
72
|
|
|
71
73
|
| Agent | Files written to the workspace |
|
|
72
74
|
|---|---|
|
|
73
75
|
| `claude-code` | `.claude/commands/setup-agent-fleet.md`, `.claude/commands/doctor-agent-fleet.md`, `.claude/skills/guided-workspace-setup/SKILL.md` |
|
|
74
|
-
| `pi` | `.pi/prompts/setup-agent-fleet.md`, `.pi/prompts/doctor-agent-fleet.md`, `.pi/skills/guided-workspace-setup/SKILL.md` |
|
|
76
|
+
| `pi` | `.pi/prompts/af-setup-agent-fleet.md`, `.pi/prompts/af-doctor-agent-fleet.md`, `.pi/skills/guided-workspace-setup/SKILL.md` |
|
|
75
77
|
| `opencode` | `.opencode/commands/af-setup-agent-fleet.md`, `.opencode/commands/af-doctor-agent-fleet.md`, `.opencode/skills/guided-workspace-setup/SKILL.md` |
|
|
76
78
|
|
|
77
79
|
These are **just the plumbing** — the slash commands, plus the skill they
|
|
78
80
|
invoke. The actual catalogue (spec-driven-development, code-reviewer,
|
|
79
|
-
test-engineer, pi extensions, …) is picked by you inside
|
|
81
|
+
test-engineer, pi extensions, …) is picked by you inside the setup workflow. Re-run
|
|
80
82
|
`init` to refresh the plumbing after a package upgrade; bootstrap files
|
|
81
83
|
are always overwritten because they're scaffolding, not user data.
|
|
82
84
|
|
|
83
|
-
After
|
|
85
|
+
After the setup workflow finishes its install pass, **the bootstrap files
|
|
84
86
|
are removed by default** so they don't clutter your agent's slash-command
|
|
85
|
-
list. Re-run `npx @chankov/agent-fleet init` whenever you want
|
|
86
|
-
|
|
87
|
+
list. Re-run `npx @chankov/agent-fleet init` whenever you want the runtime's
|
|
88
|
+
setup command back. To keep them in place across runs, reply `keep`
|
|
87
89
|
to the Step 9 confirmation prompt — the skill will record
|
|
88
90
|
`keep-installer: true` in `.ai/agent-fleet-setup.md`.
|
|
89
91
|
|
|
@@ -102,7 +104,7 @@ This is the **authoritative** record of where the npm package lives:
|
|
|
102
104
|
}
|
|
103
105
|
```
|
|
104
106
|
|
|
105
|
-
When
|
|
107
|
+
When the Agent Fleet setup command runs inside your agent, it reads this marker
|
|
106
108
|
*first* to find the source package. This matters on dev machines where you
|
|
107
109
|
may also have a git clone of `agent-fleet` elsewhere — the marker prevents
|
|
108
110
|
the skill from accidentally using that clone instead of the version the
|
|
@@ -133,7 +135,7 @@ npx @chankov/agent-fleet init --workspace ~/projects/foo --method symlink
|
|
|
133
135
|
|
|
134
136
|
Deterministic preflight scan — walks every install-target directory, lists
|
|
135
137
|
broken symlinks and stale persona references, and offers fixes. Same scan
|
|
136
|
-
that
|
|
138
|
+
that the runtime's Agent Fleet doctor command runs inside the agent.
|
|
137
139
|
|
|
138
140
|
| Flag | Default | Purpose |
|
|
139
141
|
|------|---------|---------|
|
|
@@ -150,11 +152,11 @@ npx @chankov/agent-fleet doctor -y
|
|
|
150
152
|
|
|
151
153
|
Reads the workspace's `.ai/agent-fleet-setup.md`, compares the recorded
|
|
152
154
|
package version against the installed package version, and **re-installs the
|
|
153
|
-
|
|
155
|
+
runtime's setup command** so it is always available after an update.
|
|
154
156
|
`guided-workspace-setup` removes the installer command at the end of a run by
|
|
155
157
|
default (Step 10b), so a workspace that has completed setup once would
|
|
156
158
|
otherwise have no command to hand off to. The actual diff-aware refresh then
|
|
157
|
-
runs inside the coding agent via
|
|
159
|
+
runs inside the coding agent via that command.
|
|
158
160
|
|
|
159
161
|
The agent and install method are recovered from the bootstrap marker written
|
|
160
162
|
at init time; if that marker was cleaned up too, `update` auto-detects the
|
|
@@ -165,8 +167,7 @@ Override with `--agent` / `--method`, or preview with `--dry-run`.
|
|
|
165
167
|
# Upgrade the package itself first, then check the delta:
|
|
166
168
|
npm install -g @chankov/agent-fleet@latest
|
|
167
169
|
npx agent-fleet update --workspace .
|
|
168
|
-
#
|
|
169
|
-
# it to review per-artifact diffs.
|
|
170
|
+
# The setup command is now back — run the command printed by update to review diffs.
|
|
170
171
|
```
|
|
171
172
|
|
|
172
173
|
### `npx @chankov/agent-fleet transform-persona`
|
|
@@ -177,8 +178,7 @@ Generates per-agent subagent definitions from the canonical personas in
|
|
|
177
178
|
vocabulary (`read→Read`, `find/ls→Glob`, `claude-opus-*→opus`, …; `mode:
|
|
178
179
|
subagent` + tool denials for OpenCode), agent-hub-only frontmatter dropped,
|
|
179
180
|
body untouched. pi-only personas (`bowser`, `orchestrator`) are refused for
|
|
180
|
-
other agents. This is what
|
|
181
|
-
`/setup-agent-fleet` runs during apply; transformed installs are always
|
|
181
|
+
other agents. This is what the Agent Fleet setup workflow runs during apply; transformed installs are always
|
|
182
182
|
copies (never symlinks), recorded with `transformed: true`.
|
|
183
183
|
|
|
184
184
|
| Flag | Default | Purpose |
|
|
@@ -245,7 +245,7 @@ npm is the recommended path for most users. The other two stay supported:
|
|
|
245
245
|
- **[Claude Code plugin marketplace](../README.md#quick-start)** — best UX
|
|
246
246
|
inside Claude Code. Same skills, marketplace-managed updates.
|
|
247
247
|
- **Git clone + symlinks** — best for skill authors and contributors. Clone
|
|
248
|
-
the repo, run
|
|
248
|
+
the repo, run the setup command for your runtime, and choose `symlink` in Step 8. Updates
|
|
249
249
|
flow through `git pull`. Symlinks need Developer Mode on Windows.
|
|
250
250
|
|
|
251
251
|
All three paths converge on the same `guided-workspace-setup` skill — the
|
|
@@ -261,7 +261,7 @@ npx --yes @chankov/agent-fleet@latest init --agent claude-code --method copy --w
|
|
|
261
261
|
```
|
|
262
262
|
|
|
263
263
|
`doctor` accepts `--yes` for non-interactive repair. Note that the
|
|
264
|
-
LLM-driven
|
|
264
|
+
LLM-driven setup flow is not CI-runnable by design — confirmation gates
|
|
265
265
|
exist precisely so a human approves every write.
|
|
266
266
|
|
|
267
267
|
## Receiving update notifications
|
|
@@ -298,7 +298,7 @@ is available; want me to apply it via `/setup-agent-fleet`?"*
|
|
|
298
298
|
|
|
299
299
|
### 3. pi extension (`agent-fleet-update-check`)
|
|
300
300
|
|
|
301
|
-
When installed
|
|
301
|
+
When installed from the Pi setup flow, the extension fires on the
|
|
302
302
|
first `agent_start` event of each pi session and emits a `ctx.ui.notify`
|
|
303
303
|
message in the pi UI if a newer version is published. Reads the same cache
|
|
304
304
|
as the CLI — no double-fetching.
|
package/docs/pi-extensions.md
CHANGED
|
@@ -26,7 +26,7 @@ conventions. Runtime design specs for the imported harnesses live in `docs/pi-sp
|
|
|
26
26
|
`chrome-devtools-mcp`, `compact-and-continue`, `agent-fleet-update-check`, `btw`, and
|
|
27
27
|
`pi-voice-stt`. pi
|
|
28
28
|
auto-discovers that directory, so they layer onto every session. `btw` adds a
|
|
29
|
-
`/btw <task>` prompt command (plus an `Alt+'` shortcut) that forks the current session
|
|
29
|
+
`/af-btw <task>` prompt command (plus an `Alt+'` shortcut) that forks the current session
|
|
30
30
|
into an in-process sub-session and opens a live modal over it — full context, same
|
|
31
31
|
cwd, follow-up composer, with a compact result card landing in the main transcript at
|
|
32
32
|
idle. See [.pi/extensions/btw/README.md](../.pi/extensions/btw/README.md).
|
|
@@ -37,7 +37,7 @@ Azure Speech (REST short-audio, with optional per-phrase language identification
|
|
|
37
37
|
OpenAI Whisper (Azure AI Foundry). Alt+S toggles record→insert; Enter-while-recording
|
|
38
38
|
transcribes and sends; Esc cancels. It is **gated**: the hotkey binds only once a provider is
|
|
39
39
|
configured (project-local `.ai/stt.json`, the global `~/.pi/agent/stt.json`, or `PI_STT_CONFIG`),
|
|
40
|
-
so an unconfigured session is a no-op. `/stt doctor`
|
|
40
|
+
so an unconfigured session is a no-op. `/af-stt doctor`
|
|
41
41
|
checks the setup. A simplified port of [`cgarrot/pi-voice-stt`](https://github.com/cgarrot/pi-voice-stt);
|
|
42
42
|
see [.pi/extensions/pi-voice-stt/README.md](../.pi/extensions/pi-voice-stt/README.md).
|
|
43
43
|
|
|
@@ -92,7 +92,7 @@ inside the package. The `@mariozechner/pi-*` packages are provided by the pi run
|
|
|
92
92
|
|
|
93
93
|
| Extension | Category | What it does | Run |
|
|
94
94
|
|-----------|----------|--------------|-----|
|
|
95
|
-
| [agent-hub](../.pi/harnesses/agent-hub/README.md) | Orchestration | Supported multi-agent hub: Fleet Core guardrails, dispatcher grid, specialist delegation, research helpers, persona gate, embedded coms, `/handoff`, peer-as-subagent, and footer `v<version> · <model><thinking> · <team>` — plus, inside a [herdr](https://herdr.dev) pane, fleet tools (`herdr_spawn_peer` / `herdr_read_pane` / `herdr_close_pane` with human confirmation / `herdr_notify`) | `just fleet hub` |
|
|
95
|
+
| [agent-hub](../.pi/harnesses/agent-hub/README.md) | Orchestration | Supported multi-agent hub: Fleet Core guardrails, dispatcher grid, specialist delegation, research helpers, persona gate, embedded coms, `/af-handoff`, peer-as-subagent, and footer `v<version> · <model><thinking> · <team>` — plus, inside a [herdr](https://herdr.dev) pane, fleet tools (`herdr_spawn_peer` / `herdr_read_pane` / `herdr_close_pane` with human confirmation / `herdr_notify`) | `just fleet hub` |
|
|
96
96
|
| [ask-user-remote](../.pi/harnesses/ask-user-remote/README.md) | Orchestration | Captures stock `pi-ask-user` and registers the default `ask_user`; with `user-remote` live it races local UI against the Hermes bridge, otherwise it is stock local behavior | Fleet Core (`just fleet`) |
|
|
97
97
|
| [damage-control-continue](../.pi/harnesses/damage-control-continue/README.md) | Safety | Fleet Core safety harness; guards default Pi, Hub, native children, and Herdr Pi peers. Blocks feed back without aborting the turn; protected paths support explicit approval, while dangerous command patterns remain non-exemptible | `just fleet` |
|
|
98
98
|
| [coms](../.pi/harnesses/coms/README.md) | Messaging | Peer-to-peer messaging between agents on one machine, composed with Fleet Core safety and utilities | `just fleet peer <name>` |
|
|
@@ -139,7 +139,7 @@ harnesses:
|
|
|
139
139
|
- **Dispatcher grid** — fixed specialists from `.pi/agents/teams.yaml`, shown in a live dashboard
|
|
140
140
|
with compact/full view toggling.
|
|
141
141
|
- **Specialist delegation** — `dispatch_agent` for writable child-agent work and
|
|
142
|
-
`spawn_research` / `/research` for read-only investigation.
|
|
142
|
+
`spawn_research` / `/af-research` for read-only investigation.
|
|
143
143
|
- **Verification Contract** — the dispatcher owns a ledger of checkable acceptance assertions
|
|
144
144
|
built before any builder runs, via the `set_assertions` / `update_assertion` tools. Each
|
|
145
145
|
assertion is tagged (`test` | `runtime-ui` | `code-grep` | `manual`) and advanced only on
|
|
@@ -152,20 +152,20 @@ harnesses:
|
|
|
152
152
|
dispatch refusal).
|
|
153
153
|
- **Persona gate** — requires an orchestrator persona at startup unless disabled in the local
|
|
154
154
|
override file; the chosen persona also feeds the coms purpose when no explicit `--purpose` is set.
|
|
155
|
-
- **Operator controls** — `/zoom` timeline inspection plus child-agent kill/restart controls.
|
|
155
|
+
- **Operator controls** — `/af-zoom` timeline inspection plus child-agent kill/restart controls.
|
|
156
156
|
- **Damage-control + ask_user by default** — `just fleet hub` and `just fleet hub --solo` load the
|
|
157
157
|
`damage-control-continue` safety harness and `ask-user-remote` before `agent-hub`, so the
|
|
158
158
|
dispatcher's tool calls are checked against the rules file and the `askUserAvailable` probe sees
|
|
159
159
|
`ask_user`. A blocked call feeds back and the turn keeps going rather than aborting. `agent-hub` also
|
|
160
160
|
re-loads `damage-control-continue` into every native specialist, research helper, and nested
|
|
161
161
|
delegate (via an explicit `-e` that survives `--no-extensions`). Protected-path access can be
|
|
162
|
-
granted at runtime: `/allow <pattern> [turn|session]` in the hub pre-authorizes, a parent-session
|
|
162
|
+
granted at runtime: `/af-allow <pattern> [turn|session]` in the hub pre-authorizes, a parent-session
|
|
163
163
|
block opens an approval dialog, and a headless child escalates to the hub. Protected deletion
|
|
164
164
|
offers only deny/once; other protected paths may be scoped to an agent or session. Denial and
|
|
165
165
|
timeout fail closed without aborting the child turn. Destructive bash patterns are never exemptible,
|
|
166
166
|
and a missing continue harness refuses child dispatch.
|
|
167
167
|
- **Embedded coms** — peer discovery, `coms_list` / `coms_send` / `coms_get` / `coms_await`,
|
|
168
|
-
`/handoff`, and peer-as-subagent flows.
|
|
168
|
+
`/af-handoff`, and peer-as-subagent flows.
|
|
169
169
|
- **Solo mode** — `just fleet hub --solo` keeps the dispatcher grid, delegation, research helpers, persona
|
|
170
170
|
gate, and controls, but starts without the embedded coms layer.
|
|
171
171
|
- **Optional Hermes local monitor transport** — uses `AGENT_FLEET_PROFILE_ID`, the absolute
|
package/docs/pi-setup.md
CHANGED
|
@@ -2,9 +2,9 @@
|
|
|
2
2
|
|
|
3
3
|
This guide explains how to use Agent Fleet with [pi](https://github.com/badlogic/pi-mono/tree/main/packages/coding-agent) — the terminal coding agent from `pi-mono`. Unlike some harnesses, pi has a **native Agent Fleet implementation**, so no prompt hacks are needed for skills: this repo drops in directly. Optional `pi-codex-image-gen` support is credited to <https://github.com/jvm/pi-mono>.
|
|
4
4
|
|
|
5
|
-
This repo also ships pi-native **prompt templates** for the lifecycle slash commands (`/spec`, `/plan`, `/build`, `/test`, `/review`, `/code-simplify`, `/ship`). These commands add workflow orchestration on top of the underlying skills.
|
|
5
|
+
This repo also ships pi-native **prompt templates** for the lifecycle slash commands (`/af-spec`, `/af-plan`, `/af-build`, `/af-test`, `/af-review`, `/af-code-simplify`, `/af-ship`). These commands add workflow orchestration on top of the underlying skills.
|
|
6
6
|
|
|
7
|
-
The specialist **personas** in `agents/` — including `web-performance-auditor` — are available to pi as well and are invoked directly as subagents. Note that `/webperf`
|
|
7
|
+
The specialist **personas** in `agents/` — including `web-performance-auditor` — are available to pi as well and are invoked directly as subagents. Note that `/webperf` on Claude Code (`/af-webperf` on OpenCode) does not have a Pi `.pi/prompts` equivalent; on pi you run the same audit by invoking the `web-performance-auditor` persona.
|
|
8
8
|
|
|
9
9
|
## Overview
|
|
10
10
|
|
|
@@ -21,10 +21,14 @@ This means you get near-parity with Claude Code:
|
|
|
21
21
|
- Skills are selected automatically based on intent
|
|
22
22
|
- Workflows are enforced via `AGENTS.md`
|
|
23
23
|
- Users can explicitly trigger any skill with `/skill:<name>`
|
|
24
|
-
- Users can start lifecycle workflows with `/spec`, `/plan`, `/build`, `/test`, `/review`, `/code-simplify`, and `/ship`
|
|
24
|
+
- Users can start lifecycle workflows with `/af-spec`, `/af-plan`, `/af-build`, `/af-test`, `/af-review`, `/af-code-simplify`, and `/af-ship`
|
|
25
25
|
|
|
26
26
|
No plugin, wrapper, or custom system prompt is required for the core workflow.
|
|
27
27
|
|
|
28
|
+
### Slash-command namespace
|
|
29
|
+
|
|
30
|
+
Agent Fleet owns the lowercase `/af-*` namespace on Pi. This keeps its prompt-template and harness commands distinct from Pi built-ins such as `/model` and `/settings`, skill invocations under `/skill:<name>`, and commands contributed by other installed packages. Examples: `/af-spec`, `/af-agents-list`, and `/af-allow`.
|
|
31
|
+
|
|
28
32
|
**Recommended companion packages:** [`pi-ask-user`](https://github.com/edlsh/pi-ask-user) adds an interactive `ask_user` tool and bundles an `ask-user` skill. It is bundled automatically when you install `@chankov/agent-fleet` as a pi package; clone/symlink setups should install it separately. `pi-codex-image-gen` is an optional suggested npm/pi extension for image generation; guided setup can offer it when package installation is available, but it is not bundled or required.
|
|
29
33
|
|
|
30
34
|
**Experimental phone conductor:** Linux users may pair Codex CLI `0.144.x` with ChatGPT Android to perform human-confirmed outbound delegation to live coms peers. Guided setup installs the repository assets when pi harnesses are selected, but deliberately never changes systemd, Codex auth/config, pairing, or service state. Host setup remains an explicit operator flow; see [Codex Remote-Control Conductor](codex-remote-conductor.md).
|
|
@@ -84,25 +88,25 @@ ln -s /path/to/agent-fleet/.pi/prompts .pi/prompts
|
|
|
84
88
|
This exposes:
|
|
85
89
|
|
|
86
90
|
```text
|
|
87
|
-
/spec
|
|
88
|
-
/plan
|
|
89
|
-
/build
|
|
90
|
-
/test
|
|
91
|
-
/review
|
|
92
|
-
/code-simplify
|
|
93
|
-
/ship
|
|
91
|
+
/af-spec
|
|
92
|
+
/af-plan
|
|
93
|
+
/af-build
|
|
94
|
+
/af-test
|
|
95
|
+
/af-review
|
|
96
|
+
/af-code-simplify
|
|
97
|
+
/af-ship
|
|
94
98
|
```
|
|
95
99
|
|
|
96
100
|
If `.pi/prompts` already exists as a real directory, keep it and symlink the individual command files instead:
|
|
97
101
|
|
|
98
102
|
```bash
|
|
99
|
-
ln -s /path/to/agent-fleet/.pi/prompts/spec.md .pi/prompts/spec.md
|
|
100
|
-
ln -s /path/to/agent-fleet/.pi/prompts/plan.md .pi/prompts/plan.md
|
|
101
|
-
ln -s /path/to/agent-fleet/.pi/prompts/build.md .pi/prompts/build.md
|
|
102
|
-
ln -s /path/to/agent-fleet/.pi/prompts/test.md .pi/prompts/test.md
|
|
103
|
-
ln -s /path/to/agent-fleet/.pi/prompts/review.md .pi/prompts/review.md
|
|
104
|
-
ln -s /path/to/agent-fleet/.pi/prompts/code-simplify.md .pi/prompts/code-simplify.md
|
|
105
|
-
ln -s /path/to/agent-fleet/.pi/prompts/ship.md .pi/prompts/ship.md
|
|
103
|
+
ln -s /path/to/agent-fleet/.pi/prompts/af-spec.md .pi/prompts/af-spec.md
|
|
104
|
+
ln -s /path/to/agent-fleet/.pi/prompts/af-plan.md .pi/prompts/af-plan.md
|
|
105
|
+
ln -s /path/to/agent-fleet/.pi/prompts/af-build.md .pi/prompts/af-build.md
|
|
106
|
+
ln -s /path/to/agent-fleet/.pi/prompts/af-test.md .pi/prompts/af-test.md
|
|
107
|
+
ln -s /path/to/agent-fleet/.pi/prompts/af-review.md .pi/prompts/af-review.md
|
|
108
|
+
ln -s /path/to/agent-fleet/.pi/prompts/af-code-simplify.md .pi/prompts/af-code-simplify.md
|
|
109
|
+
ln -s /path/to/agent-fleet/.pi/prompts/af-ship.md .pi/prompts/af-ship.md
|
|
106
110
|
```
|
|
107
111
|
|
|
108
112
|
4. Install the recommended `pi-ask-user` pi package separately (clone/symlink setup only):
|
|
@@ -135,7 +139,7 @@ pi
|
|
|
135
139
|
|
|
136
140
|
# then type:
|
|
137
141
|
/
|
|
138
|
-
# pi should autocomplete /spec, /plan, /build, /test, /review, /code-simplify, and /ship
|
|
142
|
+
# pi should autocomplete /af-spec, /af-plan, /af-build, /af-test, /af-review, /af-code-simplify, and /af-ship
|
|
139
143
|
```
|
|
140
144
|
|
|
141
145
|
That's it. `AGENTS.md` is already at the repo root and is auto-loaded when pi starts.
|
|
@@ -148,9 +152,9 @@ The always-on utilities:
|
|
|
148
152
|
|
|
149
153
|
- `mcp-bridge/` — a reusable factory that turns any stdio MCP server into a pi extension. This is a library consumed by wrapper extensions. Symlink it alongside wrappers so relative imports resolve; when pi discovers it directly, it intentionally registers no tools or commands by itself.
|
|
150
154
|
- `chrome-devtools-mcp/` — bridges the [`chrome-devtools-mcp`](https://www.npmjs.com/package/chrome-devtools-mcp) server into pi as native tools, unlocking the `browser-testing-with-devtools` skill on pi.
|
|
151
|
-
- `compact-and-continue/` — registers the `request_compaction` tool that queues pi context compaction to run after the current agent turn ends, optionally resuming work from a self-contained continuation prompt. Used by `/build` to offer a "Compact & continue" option at slice-approval time.
|
|
155
|
+
- `compact-and-continue/` — registers the `request_compaction` tool that queues pi context compaction to run after the current agent turn ends, optionally resuming work from a self-contained continuation prompt. Used by `/af-build` to offer a "Compact & continue" option at slice-approval time.
|
|
152
156
|
- `agent-fleet-update-check/` — surfaces an "update available" banner once per session when `@chankov/agent-fleet` has a newer published version than the one recorded in `.ai/agent-fleet-setup.md`. Never blocks startup (soft 3s check); honors `AGENT_SKILLS_NO_UPDATE_CHECK` / `NO_UPDATE_NOTIFIER` / `CI` opt-outs.
|
|
153
|
-
- `btw/` — adds the `/btw <task>` prompt command (and `Alt+'` shortcut): forks the current session into an in-process sub-session that inherits the full conversation as context, runs in the same cwd, and streams into a live modal with a follow-up composer. A compact result card lands in the main transcript at idle (kept out of the main agent's LLM context). See [.pi/extensions/btw/README.md](../.pi/extensions/btw/README.md).
|
|
157
|
+
- `btw/` — adds the `/af-btw <task>` prompt command (and `Alt+'` shortcut): forks the current session into an in-process sub-session that inherits the full conversation as context, runs in the same cwd, and streams into a live modal with a follow-up composer. A compact result card lands in the main transcript at idle (kept out of the main agent's LLM context). See [.pi/extensions/btw/README.md](../.pi/extensions/btw/README.md).
|
|
154
158
|
|
|
155
159
|
To install, symlink the directories into your project's `.pi/extensions/`:
|
|
156
160
|
|
|
@@ -173,7 +177,7 @@ npm ci
|
|
|
173
177
|
|
|
174
178
|
Because the project extensions are symlinks into the clone, these dependencies are reused by every project that links the same extension directories.
|
|
175
179
|
|
|
176
|
-
Verify by starting `pi` and running `/chrome_devtools-status` — expect `Chrome DevTools MCP connected. Registered N tool(s).`
|
|
180
|
+
Verify by starting `pi` and running `/af-chrome_devtools-status` — expect `Chrome DevTools MCP connected. Registered N tool(s).`
|
|
177
181
|
|
|
178
182
|
#### Extension harnesses — orchestration, safety, messaging
|
|
179
183
|
|
|
@@ -284,8 +288,8 @@ pi also searches for prompt templates in:
|
|
|
284
288
|
Each Markdown file becomes a slash command by filename. For example:
|
|
285
289
|
|
|
286
290
|
```
|
|
287
|
-
.pi/prompts/spec.md → /spec
|
|
288
|
-
.pi/prompts/code-simplify.md → /code-simplify
|
|
291
|
+
.pi/prompts/af-spec.md → /af-spec
|
|
292
|
+
.pi/prompts/af-code-simplify.md → /af-code-simplify
|
|
289
293
|
```
|
|
290
294
|
|
|
291
295
|
These lifecycle commands are not replacements for skills. They are workflow entry points that add orchestration and tell the agent which skills to load and follow for the current phase.
|
|
@@ -295,20 +299,20 @@ These lifecycle commands are not replacements for skills. They are workflow entr
|
|
|
295
299
|
Three ways to trigger the workflow:
|
|
296
300
|
|
|
297
301
|
- **Explicit skill:** type `/skill:<name>` (e.g. `/skill:spec-driven-development`)
|
|
298
|
-
- **Lifecycle command:** type `/spec`, `/plan`, `/build`, `/test`, `/review`, `/code-simplify`, or `/ship`
|
|
302
|
+
- **Lifecycle command:** type `/af-spec`, `/af-plan`, `/af-build`, `/af-test`, `/af-review`, `/af-code-simplify`, or `/af-ship`
|
|
299
303
|
- **Automatic:** describe intent in natural language — the model reads `AGENTS.md` and loads the matching skill
|
|
300
304
|
|
|
301
305
|
### 5. Lifecycle Mapping
|
|
302
306
|
|
|
303
307
|
The development lifecycle is encoded in both `AGENTS.md` and the pi prompt templates:
|
|
304
308
|
|
|
305
|
-
- DEFINE → `/spec` → `spec-driven-development`
|
|
306
|
-
- PLAN → `/plan` → `planning-and-task-breakdown`
|
|
307
|
-
- BUILD → `/build` → `incremental-implementation` + `test-driven-development`
|
|
308
|
-
- VERIFY → `/test` → `test-driven-development`; use `debugging-and-error-recovery` when tests or builds fail
|
|
309
|
-
- REVIEW → `/review` → `code-review-and-quality`
|
|
310
|
-
- SIMPLIFY → `/code-simplify` → `code-simplification`
|
|
311
|
-
- SHIP → `/ship` → `shipping-and-launch`
|
|
309
|
+
- DEFINE → `/af-spec` → `spec-driven-development`
|
|
310
|
+
- PLAN → `/af-plan` → `planning-and-task-breakdown`
|
|
311
|
+
- BUILD → `/af-build` → `incremental-implementation` + `test-driven-development`
|
|
312
|
+
- VERIFY → `/af-test` → `test-driven-development`; use `debugging-and-error-recovery` when tests or builds fail
|
|
313
|
+
- REVIEW → `/af-review` → `code-review-and-quality`
|
|
314
|
+
- SIMPLIFY → `/af-code-simplify` → `code-simplification`
|
|
315
|
+
- SHIP → `/af-ship` → `shipping-and-launch`
|
|
312
316
|
|
|
313
317
|
---
|
|
314
318
|
|
|
@@ -329,7 +333,7 @@ pi behavior:
|
|
|
329
333
|
|
|
330
334
|
Equivalent explicit forms:
|
|
331
335
|
```
|
|
332
|
-
/spec Add authentication to this app
|
|
336
|
+
/af-spec Add authentication to this app
|
|
333
337
|
/skill:spec-driven-development
|
|
334
338
|
```
|
|
335
339
|
|
|
@@ -348,7 +352,7 @@ pi behavior:
|
|
|
348
352
|
|
|
349
353
|
Equivalent explicit forms:
|
|
350
354
|
```
|
|
351
|
-
/test This endpoint is returning 500 errors
|
|
355
|
+
/af-test This endpoint is returning 500 errors
|
|
352
356
|
/skill:debugging-and-error-recovery
|
|
353
357
|
```
|
|
354
358
|
|
|
@@ -367,7 +371,7 @@ pi behavior:
|
|
|
367
371
|
|
|
368
372
|
Equivalent explicit forms:
|
|
369
373
|
```
|
|
370
|
-
/review
|
|
374
|
+
/af-review
|
|
371
375
|
/skill:code-review-and-quality
|
|
372
376
|
```
|
|
373
377
|
|
|
@@ -392,16 +396,16 @@ After installing, confirm the integration works:
|
|
|
392
396
|
|
|
393
397
|
1. Run `pi` from inside the repo (or any subdirectory).
|
|
394
398
|
2. Type `/skill:` and confirm the skill list autocompletes with entries like `spec-driven-development`, `incremental-implementation`, `code-review-and-quality`, and `ask-user`.
|
|
395
|
-
3. Type `/` and confirm the lifecycle commands autocomplete: `/spec`, `/plan`, `/build`, `/test`, `/review`, `/code-simplify`, and `/ship`.
|
|
396
|
-
4. Run `/spec design a new feature for X` — confirm pi expands the command and invokes `spec-driven-development`.
|
|
397
|
-
5. Ask: *"fix this bug"* — confirm pi invokes `debugging-and-error-recovery`, or run `/test` to start a TDD/debugging workflow explicitly.
|
|
399
|
+
3. Type `/` and confirm the lifecycle commands autocomplete: `/af-spec`, `/af-plan`, `/af-build`, `/af-test`, `/af-review`, `/af-code-simplify`, and `/af-ship`.
|
|
400
|
+
4. Run `/af-spec design a new feature for X` — confirm pi expands the command and invokes `spec-driven-development`.
|
|
401
|
+
5. Ask: *"fix this bug"* — confirm pi invokes `debugging-and-error-recovery`, or run `/af-test` to start a TDD/debugging workflow explicitly.
|
|
398
402
|
6. Give pi an ambiguous or high-stakes request and confirm it can use the `ask_user` tool / `ask-user` skill to request an explicit decision.
|
|
399
403
|
|
|
400
404
|
If skill autocomplete is empty, check that `.agents/skills` points to a directory containing `<skill-name>/SKILL.md` files and that pi was not started with `--no-skills`.
|
|
401
405
|
|
|
402
406
|
If lifecycle command autocomplete is empty, check that `.pi/prompts` points to a directory containing the command Markdown files and run `/reload` or restart pi.
|
|
403
407
|
|
|
404
|
-
If `/chrome_devtools-status` reports `Cannot find module '@modelcontextprotocol/sdk/client/index.js'`, the MCP extension loaded but its runtime initialization failed. An `[Extensions]` startup entry alone does not prove its tools registered.
|
|
408
|
+
If `/af-chrome_devtools-status` reports `Cannot find module '@modelcontextprotocol/sdk/client/index.js'`, the MCP extension loaded but its runtime initialization failed. An `[Extensions]` startup entry alone does not prove its tools registered.
|
|
405
409
|
|
|
406
410
|
For a cloned or copied install, run:
|
|
407
411
|
|
|
@@ -410,7 +414,7 @@ cd /path/to/agent-fleet/.pi/extensions
|
|
|
410
414
|
npm ci
|
|
411
415
|
```
|
|
412
416
|
|
|
413
|
-
For extensions symlinked into a published `@chankov/agent-fleet` package, update to a package version that declares the extension dependencies at its root, then restart/reload pi. Do not rely on installing only the target workspace's `.pi/extensions/node_modules`: Node may resolve the symlink to the package's real path. Verify the repair with `/chrome_devtools-status`; it must report `connected` and a non-zero registered tool count.
|
|
417
|
+
For extensions symlinked into a published `@chankov/agent-fleet` package, update to a package version that declares the extension dependencies at its root, then restart/reload pi. Do not rely on installing only the target workspace's `.pi/extensions/node_modules`: Node may resolve the symlink to the package's real path. Verify the repair with `/af-chrome_devtools-status`; it must report `connected` and a non-zero registered tool count.
|
|
414
418
|
|
|
415
419
|
The harnesses install separately — if a harness reports `Cannot find module 'yaml'` or `'@sinclair/typebox'`, run `npm ci` in `.pi/harnesses/` as well (or `just install` from the clone, which does both).
|
|
416
420
|
|
|
@@ -439,13 +443,13 @@ Just use natural language:
|
|
|
439
443
|
|
|
440
444
|
Or invoke lifecycle commands when you want the full workflow prompt:
|
|
441
445
|
|
|
442
|
-
- `/spec`
|
|
443
|
-
- `/plan`
|
|
444
|
-
- `/build`
|
|
445
|
-
- `/test`
|
|
446
|
-
- `/review`
|
|
447
|
-
- `/code-simplify`
|
|
448
|
-
- `/ship`
|
|
446
|
+
- `/af-spec`
|
|
447
|
+
- `/af-plan`
|
|
448
|
+
- `/af-build`
|
|
449
|
+
- `/af-test`
|
|
450
|
+
- `/af-review`
|
|
451
|
+
- `/af-code-simplify`
|
|
452
|
+
- `/af-ship`
|
|
449
453
|
|
|
450
454
|
Or invoke individual skills directly when you want precise control:
|
|
451
455
|
|
|
@@ -465,7 +469,7 @@ pi integration works by leveraging pi's **native** Agent Fleet and prompt-templa
|
|
|
465
469
|
- Symlink `.pi/prompts/` into the target project's `.pi/prompts/`
|
|
466
470
|
- Install `@chankov/agent-fleet` as a pi package for bundled `ask_user`, or install `pi-ask-user` separately for clone/symlink setup
|
|
467
471
|
- Let pi auto-load `AGENTS.md` from the repo root
|
|
468
|
-
- Use `/skill:<name>`, lifecycle commands like `/spec`, or natural language to trigger workflows
|
|
472
|
+
- Use `/skill:<name>`, lifecycle commands like `/af-spec`, or natural language to trigger workflows
|
|
469
473
|
|
|
470
474
|
Guided setup can additionally offer the optional `pi-codex-image-gen` package when pi package installation is available, but it is not part of the minimal path.
|
|
471
475
|
|
package/hermes/README.md
CHANGED
|
@@ -20,7 +20,7 @@ agent-fleet set-hermes-telegram install --profile default --force
|
|
|
20
20
|
agent-fleet set-hermes-telegram install --profile default --force --restart
|
|
21
21
|
```
|
|
22
22
|
|
|
23
|
-
|
|
23
|
+
Pi and OpenCode use `/af-set-hermes-telegram`; Claude Code uses `/set-hermes-telegram`. The installer resolves the real profile path via `hermes profile show`, refuses drift without `--force`, backs up forced replacements, copies atomically, and verifies skill/tool/gateway readiness. It never sends a test Telegram message and never starts a stopped gateway. See [the bridge runbook](../docs/coms-hermes-bridge.md#set-hermes-telegram-bridge-control).
|
|
24
24
|
|
|
25
25
|
The current `hermes skills install` CLI documents registry identifiers and direct HTTP(S) `SKILL.md` URLs, not local directories, so `hermes skills install hermes/skills/hub-liaison` is not the supported local installation path.
|
|
26
26
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@chankov/agent-fleet",
|
|
3
|
-
"version": "0.0.
|
|
3
|
+
"version": "0.0.6",
|
|
4
4
|
"description": "Multi-agent fleet orchestration for coding agents. Pi as the primary runtime, Herdr as the workspace control plane, coms as the peer data plane, and Hermes for remote control — plus skills, personas, harnesses, and a guided setup CLI.",
|
|
5
5
|
"keywords": [
|
|
6
6
|
"ai",
|
|
@@ -87,7 +87,7 @@ defined behavior on partial arrival.
|
|
|
87
87
|
blocks on the `msg_id`, the receiver's `agent_end` response capture releases it.
|
|
88
88
|
- The agent-hub dispatcher awaiting parallel dispatches before the Verification
|
|
89
89
|
Contract's gate — no verdict until every specialist's structured return is in.
|
|
90
|
-
- `/handoff` — a full-session barrier: the dispatcher composes a brief, sends it, and
|
|
90
|
+
- `/af-handoff` — a full-session barrier: the dispatcher composes a brief, sends it, and
|
|
91
91
|
awaits the peer's acknowledgment before ceding.
|
|
92
92
|
|
|
93
93
|
**Failure modes:** deadlock by mutual await (A awaits B while B awaits A — the coms hop
|
|
@@ -132,7 +132,7 @@ validation; every other runner is torn down immediately. Requires real kill sema
|
|
|
132
132
|
and a validation step *before* declaring a winner (first ≠ correct).
|
|
133
133
|
|
|
134
134
|
**In this repo:** **not implemented as a first-class mechanism.** The pieces exist —
|
|
135
|
-
herdr `pane.close` / `workspace.close` (and the agent-hub `/agents-kill` command) can
|
|
135
|
+
herdr `pane.close` / `workspace.close` (and the agent-hub `/af-agents-kill` command) can
|
|
136
136
|
tear down runners, and `coms_get` can poll multiple pending `msg_id`s — but **coms has no
|
|
137
137
|
cancel envelope**: a losing peer keeps working (and burning tokens) until its turn ends;
|
|
138
138
|
its late response is dropped as an orphan. Racing today means paying for every lane.
|
|
@@ -75,7 +75,7 @@ ln -s /path/to/agent-fleet/.pi/extensions/mcp-bridge .pi/extensions/mcp
|
|
|
75
75
|
ln -s /path/to/agent-fleet/.pi/extensions/chrome-devtools-mcp .pi/extensions/chrome-devtools-mcp
|
|
76
76
|
```
|
|
77
77
|
|
|
78
|
-
After install, run `/
|
|
78
|
+
After install, run `/af-chrome-devtools-status` inside pi to confirm the bridge connected.
|
|
79
79
|
|
|
80
80
|
### Available Tools
|
|
81
81
|
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: compound-learning
|
|
3
|
-
description: Captures a finished session's lessons into the project's own rules and docs — extracts candidate lessons from session evidence, dedupes them against the existing rule tree, and lands them as minimal diffs on existing files. Use when a session ends with user corrections, recurring review findings, or debugging root causes worth keeping, when the user says "compound" or "capture the lessons", or as the documenter's process for a /compound dispatch.
|
|
3
|
+
description: Captures a finished session's lessons into the project's own rules and docs — extracts candidate lessons from session evidence, dedupes them against the existing rule tree, and lands them as minimal diffs on existing files. Use when a session ends with user corrections, recurring review findings, or debugging root causes worth keeping, when the user says "compound" or "capture the lessons", or as the documenter's process for a /compound or /af-compound dispatch.
|
|
4
4
|
---
|
|
5
5
|
|
|
6
6
|
# Compound Learning
|
|
@@ -16,7 +16,7 @@ The enemy is not missing lessons — it is **bloat**. A rule tree that grows by
|
|
|
16
16
|
## When to Use
|
|
17
17
|
|
|
18
18
|
- A session is wrapping up and it contained at least one correction, recurring finding, or hard-won discovery.
|
|
19
|
-
- The user says "compound", "capture the lessons", "update the rules from this session", or runs a `/compound` command.
|
|
19
|
+
- The user says "compound", "capture the lessons", "update the rules from this session", or runs a `/compound` or `/af-compound` command.
|
|
20
20
|
- You are the `documenter` persona dispatched with a compound task by the agent-hub dispatcher.
|
|
21
21
|
- Periodically, as a **consolidation pass** over a rule tree that has accumulated additions (see Step 8).
|
|
22
22
|
|
|
@@ -32,7 +32,7 @@ The enemy is not missing lessons — it is **bloat**. A rule tree that grows by
|
|
|
32
32
|
| Session diff | `git diff` / `git log` for the session's commits |
|
|
33
33
|
| agent-hub artifacts | `.pi/agent-sessions/artifacts/{returns,reviews,plans,inventories,evidence}/` — specialist returns and review findings |
|
|
34
34
|
| agent-hub assertion ledger | `.pi/agent-sessions/assertions.json` — which assertions failed first and why |
|
|
35
|
-
| Dispatch brief | A `/compound` dispatch carries a candidate-lessons brief composed by the dispatcher — treat it as *candidates*, not conclusions |
|
|
35
|
+
| Dispatch brief | A `/compound` or `/af-compound` dispatch carries a candidate-lessons brief composed by the dispatcher — treat it as *candidates*, not conclusions |
|
|
36
36
|
|
|
37
37
|
**Targets** — from `.ai/agent-fleet-overrides.md`, section `## agent-hub` (legacy `## agent-team`):
|
|
38
38
|
|