By name: autonomous-pipeline
pipeline_version: "6.0" # 3 gates (Gate 0 EVALUATE family + Gate 1 Skeptic/SSA + Gate 2 Adversarial) · 10 stages · 2 execution modes (full one-shot / goal iterative). See INSTRUCTIONS.md Architecture.
description: "Orchestrate the full AIDLC Autonomous Pipeline from a one-sentence requirement to a PR-ready
\ delivery with TDD methodology. DDD drives judgment, SDD produces specs, TDD verifies delivery.
\ Stages: Evaluate, Think, Plan, Build, Review, Test, Deliver, Reflect. Checkpoints on BLOCK or context limits.\n
\ TRIGGER: "run pipeline", "autonomous pipeline", "pipeline for", "full pipeline".\n
\ NOT FOR: single stages (use evaluate, deep-research, code-review, qa, deliver directly)."
consumes_artifacts:
- evaluation
- research
- alternatives
- design_doc
- changeset
- review
- test_report produces_artifact:
- evaluation
- research
- design_doc
- changeset
- review
- test_report
- delivery
- checkpoint tier: lazy
Autonomous Pipeline
This skill loads full instructions on activation. Read INSTRUCTIONS.md before proceeding.
TRIGGER: "run pipeline", "autonomous pipeline", "pipeline for", "full pipeline", DO NOT USE: for a single stage (use the specific skill: evaluate, deep-research,
Verification
Before marking the pipeline complete, show evidence for each:
- REPORT.md generated — saved to
.artifacts/runs/<RUN_ID>/REPORT.md - All stage artifacts published — every completed stage has a published artifact
- Confidence score calculated — score breakdown shown, not just a number
- Decision log complete — every non-trivial decision classified (mechanical/taste/judgment)
- TDD cycle verified — RED (tests fail) → GREEN (tests pass) → VERIFY (no regressions)