-
Notifications
You must be signed in to change notification settings - Fork 0
Project Index
Nick edited this page Jun 20, 2026
·
6 revisions
Generated from the current repository code on 2026-06-11 by npm run sync:project-index.
- Package:
cool-workflow - Version:
0.1.78 - Source modules:
58 - Workflow apps:
6 - Docs:
46 - Smoke tests:
69 - Repository: https://github.com/coo1white/cool-workflow
workflow app -> runner -> dispatch -> isolated workers
-> results -> feedback/candidates -> verifier gate
-> commit/checkpoint -> report/trust audit
multi-agent host -> topology -> blackboard/coordinator
-> fanout/fanin -> candidate score/select
| Module | Responsibility |
|---|---|
| orchestrator.ts | Plans runs, loads workflows, records results, writes reports, and exposes runner commands. |
| state.ts | Persists run checkpoints, JSON state, run paths, and state migration entrypoints. |
| state-node.ts | Defines explicit state nodes, pipeline transitions, evidence checks, and node persistence. |
| pipeline-contract.ts | Builds the default pipeline contract used by run state. |
| pipeline-runner.ts | Finds runnable stages and advances/fails pipeline nodes with retry-aware errors. |
| types.ts | Owns the shared workflow, run, app, evidence, worker, candidate, audit, and topology types. |
| Module | Responsibility |
|---|---|
| verifier.ts | Validates result envelopes, findings, evidence, and run gate completion. |
| commit.ts | Creates verifier-gated commits and explicit manual checkpoints. |
| candidate-scoring.ts | Registers, scores, ranks, selects, rejects, and summarizes candidate outputs. |
| error-feedback.ts | Turns failures into persisted feedback records and correction tasks. |
| trust-audit.ts | Records provenance, sandbox decisions, host attestations, and acceptance rationale. |
| Module | Responsibility |
|---|---|
| dispatch.ts | Selects runnable tasks and writes dispatch manifests. |
| worker-isolation.ts | Allocates worker scopes, writes manifests, records worker outputs, and validates boundaries. |
| sandbox-profile.ts | Resolves named sandbox policy contracts and validates read/write/command/network boundaries. |
| harness.ts | Renders task files for dispatched work. |
| Module | Responsibility |
|---|---|
| multi-agent.ts | Persists multi-agent runs, roles, groups, memberships, fanouts, and fanins. |
| coordinator.ts | Owns blackboard topics, messages, context, artifacts, snapshots, and coordinator decisions. |
| topology.ts | Defines and applies official map-reduce, debate, and judge-panel topologies. |
| multi-agent-host.ts | Provides the preferred host loop for run, status, step, blackboard, score, and select. |
| Module | Responsibility |
|---|---|
| cli.ts | Routes human CLI commands to runtime, app, topology, multi-agent, and operator flows. |
| mcp-server.ts | Exposes JSON-RPC/MCP tool parity for agent hosts. |
| operator-ux.ts | Formats status, reports, graph, worker, candidate, feedback, commit, and trust summaries. |
| workflow-app-framework.ts | Validates app manifests and loads app entrypoints. |
| workflow-api.ts | Provides the fluent workflow, phase, task, artifact, and input API. |
| daemon.ts | Runs scheduled tasks through the desktop scheduler daemon. |
| scheduler.ts | Creates, stores, computes, and runs schedules. |
| triggers.ts | Bridges routine triggers to explicit workflow events. |
| version.ts | Defines current package and state schema versions. |
- agent-config.ts
- capability-core.ts
- capability-registry.ts
- collaboration.ts
- contract-migration.ts
- drive.ts
- evidence-grounding.ts
- evidence-reasoning.ts
- execution-backend.ts
- loop-expansion.ts
- multi-agent-eval.ts
- multi-agent-operator-ux.ts
- multi-agent-trust.ts
- node-snapshot.ts
- observability.ts
- reclamation.ts
- result-normalize.ts
- run-export.ts
- run-registry.ts
- run-state-schema.ts
- scheduling.ts
- schema-validate.ts
- state-explosion.ts
- state-migrations.ts
- telemetry-attestation.ts
- telemetry-demo.ts
- telemetry-ledger.ts
- verifier-registry.ts
- workbench-host.ts
- workbench.ts
| App | Type | Inputs | Sandbox | Source |
|---|---|---|---|---|
architecture-review - Map a repository architecture, assess risks, verify important findings, and synthesize an evidence-backed verdict. |
canonical |
repo, question, invariant, focus
|
readonly |
manifest / workflow |
end-to-end-golden-path - Deterministic one-worker workflow app for proving the CW integration chain. |
userland | question |
readonly |
manifest / workflow |
pr-review-fix-ci - Review a pull request or branch, inspect CI failures, diagnose actionable issues, optionally patch, verify, and summarize with evidence. |
canonical |
repo, pr, branch, base, ci, mode
|
readonly, workspace-write
|
manifest / workflow |
release-cut - Prepare a release with checklist discipline: version checks, changelog, tests, packaging, release notes, and final verification. |
canonical |
repo, version, previousVersion, releaseBranch, dryRun
|
readonly, workspace-write
|
manifest / workflow |
research-synthesis - Split a research question into claims, investigate sources, cross-check evidence, verify claims, and synthesize a concise answer. |
canonical |
question, source, scope, freshness
|
readonly, locked-down
|
manifest / workflow |
workflow-app-framework-demo - Small framework app showing inputs, phases, evidence gates, and sandbox profile hints. |
example | question |
readonly, workspace-write
|
manifest / workflow |
- Agent Delegation Drive
- Workflow App framework
- CANDIDATE-SCORING(7)
- Canonical Workflow Apps
- CAPABILITY-TOPOLOGY-REGISTRY(7) — Cool Workflow Agent-Driven Self-Evolution
- CLI ↔ MCP Parity
- Contract Migration Tooling
- Control-Plane Scheduling
- Coordinator / Blackboard
- Dogfood One Real Repo
- Durable State & Locking
- End-to-End Golden Path
- ERROR-FEEDBACK(7)
- Evidence Adoption Reasoning Chain
- EXECUTION-BACKENDS(7)
- Getting Started
- Cool Workflow Docs
- MCP App Surface
- Multi-Agent CLI + MCP Surface
- Multi-Agent Eval & Replay Harness
- Multi-Agent Operator UX
- Multi-Agent Runtime Core
- Multi-Agent Topologies
- Multi-Agent Trust / Policy / Audit
- Node Snapshot / Diff / Replay
- Observability + Cost Accounting
- Operator UX
- PIPELINE-RUNNER(7)
- Cool Workflow Project Index
- Real Execution Backend Integrations
- Release And Migration Discipline
- Release Tooling
- Routines
- Run Registry / Control Plane
- Run Retention & Provable Reclamation
- SANDBOX-PROFILES(7)
- Scheduled Tasks
- Security / Trust Hardening
- State Explosion Management
- STATE-NODE(7)
- Team Collaboration
- Unix-Inspired Workflow Principles
- VERIFIER-GATED-COMMIT(7)
- Web / Desktop Workbench
- WORKER-ISOLATION(7)
- Workflow App framework
Smoke tests mirror the public contracts. The high-signal suites are:
- agent-delegation-drive-smoke.js
- artifact-integrity-smoke.js
- backend-registry-smoke.js
- block-unapproved-tag-smoke.js
- candidate-scoring-smoke.js
- canonical-workflow-apps-smoke.js
- claude-p-agent-wrapper-smoke.js
- cli-mcp-parity-smoke.js
- concurrent-failure-semantics-smoke.js
- concurrent-workflow-dsl-smoke.js
- contract-migration-tooling-smoke.js
- control-plane-scheduling-smoke.js
- coordinator-blackboard-smoke.js
- dogfood-release-smoke.js
- durable-atomic-write-smoke.js
- end-to-end-demo-smoke.js
- end-to-end-golden-path-smoke.js
- error-feedback-resolution-smoke.js
- error-feedback-smoke.js
- evidence-adoption-reasoning-smoke.js
- evidence-content-extraction-smoke.js
- execution-backends-smoke.js
- mcp-app-surface-smoke.js
- multi-agent-cli-mcp-surface-smoke.js
- multi-agent-eval-replay-harness-smoke.js
- multi-agent-eval-replay-smoke.js
- multi-agent-operator-ux-smoke.js
- multi-agent-runtime-core-smoke.js
- multi-agent-topologies-smoke.js
- multi-agent-trust-policy-audit-smoke.js
- no-false-green-smoke.js
- node-snapshot-diff-replay-smoke.js
- observability-cost-accounting-smoke.js
- one-way-boundary-smoke.js
- operator-ux-smoke.js
- parallel-onramp-smoke.js
- pipeline-auto-advance-smoke.js
- pipeline-runner-smoke.js
- project-index-sync-smoke.js
- quickstart-smoke.js
- real-execution-backends-smoke.js
- release-flow-smoke.js
- release-gate-smoke.js
- release-tooling-smoke.js
- result-normalize-smoke.js
- robustness-hardening-smoke.js
- run-export-import-smoke.js
- run-fixture-compat-smoke.js
- run-registry-control-plane-smoke.js
- run-retention-reclamation-smoke.js
- sandbox-profile-smoke.js
- schema-validation-smoke.js
- security-trust-hardening-smoke.js
- self-audit-hardening-smoke.js
- state-explosion-management-smoke.js
- state-node-smoke.js
- tamper-evidence-demo-smoke.js
- team-collaboration-smoke.js
- telemetry-attest-wrap-smoke.js
- telemetry-attestation-smoke.js
- telemetry-fail-closed-smoke.js
- telemetry-ledger-smoke.js
- telemetry-metrics-coverage-smoke.js
- token-budget-enforcement-smoke.js
- verifier-gated-commit-smoke.js
- web-desktop-workbench-smoke.js
- worker-isolation-smoke.js
- worker-retry-count-smoke.js
- workflow-app-framework-smoke.js
- Repository docs: docs/project-index.md
- Obsidian vault (optional): set
CW_OBSIDIAN_VAULTto your local vault path. - GitHub Wiki: the
cool-workflow.wikiworking tree (override withCW_GITHUB_WIKI_DIR).
Run this after changing source modules, workflow app manifests, public docs, or smoke test coverage:
cd plugins/cool-workflow
npm run sync:project-indexThen review the Obsidian page and GitHub Wiki working tree before publishing wiki changes.
Organized from local Obsidian notes and reconciled with the current
coo1white/cool-workflow repository state.
Start here
Go deeper
- Workflow Apps
- Architecture
- Trust And Audit
- Recovery And Restore
- Commands or API
- MCP And Manifests
- Operations
- FAQ
Source docs