agentleFS
gongyijie85gongyijie85/dsh-ecc/skills/orch-pipeline/SKILL.md

orch-pipeline

Skill8 starsChanged 3 months agoMITView on GitHub

Shared orchestration engine for the orch-* skill family. Defines the gated Research-Plan-TDD-Review-Commit pipeline, the size classifier, the agent map, and the two human gates that the orch-* operation skills delegate to. Not usually invoked directly. Not usually invoked directly; it applies when an orch-* skill delegates its gated Research-Plan-TDD-Review-Commit pipeline.

3 other projects use a version of this, 3 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 1 is common, 0 are distinctive, and 0 are very specific to this project.

  • 1

    Common

    Most of the 3 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Common:

    Read directly only when adding a new operation to the family or tuning the shared phases, gates, or agent map.

    line 20

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Orchestrator Pipeline (shared engine)

The orch-* skills are thin wrappers. They do not re-implement any work — they classify the request, choose which phases of this pipeline run, and delegate each phase to an existing ECC agent or command. This file is that pipeline.

Invoke an operation skill (orch-add-feature, orch-fix-defect, …) rather than this engine directly. This file is the reference they point at.

When to Use

  • Loaded indirectly whenever an orch-* operation skill runs.
  • Read directly only when adding a new operation to the family or tuning the shared phases, gates, or agent map.

The operation family

SkillOperationTriggerFirst move
orch-add-featurefeaturecapability does not exist yetresearch + plan a new slice
orch-change-featuretweakworks, but desired behavior differsamend existing behavior and its tests
orch-fix-defectfixbroken; behavior is wrongreproduce as a failing test, then fix
orch-refine-coderefactorbehavior stays, structure improvesrestructure while keeping tests green
orch-build-mvpmvpbootstrap from a design/spec docingest doc → vertical slices

These wrappers compose existing ECC commands rather than replace them: /feature-dev, /plan, /code-review, /build-fix, /refactor-clean, and /gan-build, plus the tdd-workflow skill. The orch-* family adds the shared size classifier and the two gates on top of them, so one umbrella covers all five operations consistently.

Step 0 — Classify size (right-sizing)

Ceremony scales to blast radius. Score the request on three signals, take the highest tier any signal reaches, and state the result in one line so the user can override:

TierFiles touchedNew dependency / contractDesign ambiguityPhases that run
trivial1, a few linesnonenone — the change is obvious4 → 5 → 6
small1 file / 1 functionnoneclear once you read the code(1 light) → 4 → 5 → 6
standard2–5 filesmaybe a new internal moduleone real choice to make1 → 2 → 4 → 5 → 6
largemany / cross-cuttingnew external dep, public API, or a spec docmultiple open questions1 → 2 → (3) → 4 → 5 → 6

Phase 0 (Intake) always runs and is omitted from the mask column above. The tie-breaker: anything touching a security trigger (below) or a public API / contract is at least standard, regardless of file count.

The phases

Each phase delegates — it does not do the work inline.

  • 0. Intake — restate the request. For orch-build-mvp, read the spec/design doc and extract scope, locked decisions, and a feature list.
  • 1. Research & Reuse — per rules/common/development-workflow.md: gh search repos / gh search code, then Context7 / vendor docs, then package registries, then Exa. Prefer adopting a proven implementation over net-new code.
  • 2. Plan — delegate to the planner agent (or architect / code-architect for structural decisions). Output a task_list ordered as thin vertical slices. → GATE 1.
  • 3. Scaffold — orch-build-mvp only: stand up the first end-to-end slice.
  • 4. Implement (TDD) — drive each task through the tdd-guide agent (or the tdd-workflow skill): red → green → refactor. Honor the operation's first-move rule.
  • 5. Review — code-reviewer agent / /code-review. Add security-reviewer whenever the diff touches a security trigger (below).
  • 6. Commit — conventional commits (feat: / fix: / refactor: / …), one per logical chunk. → GATE 2.

The two gates

This family is gated, not autonomous:

  1. GATE 1 — after Plan. Present the task_list; do not write implementation code until the user approves.
  2. GATE 2 — before Commit. Present the diff summary and proposed messages; do not commit until the user confirms.

Everything between the gates flows without stopping.

Agent / command map

PhasePrimaryFallback / escalation
Intake / understandcode-explorertrace existing paths before a tweak, fix, or refactor
Planplannerarchitect, code-architect for structural calls
Implementtdd-guide (or tdd-workflow skill)build-error-resolver / /build-fix on build breaks
Reviewcode-reviewer / /code-reviewlanguage reviewer (python-reviewer, typescript-reviewer, …)
Securitysecurity-reviewer—
MVP inner loop/gan-build "<brief>" --skip-plannerdrives gan-generator → gan-evaluator; tune --max-iterations / --pass-threshold

Match the language reviewer to the repo (see the repo's own CLAUDE.md).

Security-review trigger

Pull in security-reviewer when the diff touches any of: authentication or authorization, user-input handling, database queries, file-system paths, external API calls, cryptography, or secrets / credentials. (Per rules/common/security.md.)

Handoff artifacts

The pipeline carries no hidden state — the planning docs are the handoff:

  • task_list (from Plan) drives the Implement loop.
  • Larger work may also emit PRD / architecture / system_design under the repo's docs/ per rules/common/development-workflow.md.
  • Review findings (CRITICAL / HIGH) must be resolved before Gate 2.

Verification

  • size tier was stated and matched the work
  • Gate 1 (plan) and Gate 2 (commit) were both honored
  • security-reviewer ran iff a security trigger was touched
  • commits are conventional and scoped to one logical change
  • new / changed behavior has tests; coverage ≥ 80% per rules/common/testing.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.