agentleFS
QuirkybrainQuirkybrain/claude-code-swe-framework/.claude/skills/quality-gate/SKILL.md

quality-gate

Skill1 starsChanged 53 days agoMITView on GitHub

Evaluate a Component or task against its Definition of Done with checks scaled to LIGHTWEIGHT, STANDARD, or STRICT. Use after implementation and current validation evidence exist, before marking work DONE or a Component STABLE.

42 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use deploy.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • VercelSkill

    deploy

    Deploy applications to Vercel. Use when the user says "deploy", "deploy to Vercel", "push to production", "deploy my app", or "go live".

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 34 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run after implementation, build, testing, and required review.

    line 14
  2. Distinctive:

    All evidence names the exact evaluated revision and scope.

    line 25
  3. Distinctive:

    Add checks demanded by actual impact even if the selected level baseline is lighter.

    line 36
  4. Distinctive:

    Verify every mandatory DoD item and accepted-risk authority.

    line 37
  5. Distinctive:

    Use Failure Diagnosis before repeated correction.

    line 64
  6. Very specific:

    Stop after a recorded verdict.

    line 68

“Common” is measured: kept by at least 6 in 10 of the 34 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

Quality Gate

Purpose

Issue an evidence-based Component Quality Gate verdict appropriate to the effective Task Workflow Level.

Trigger

Run after implementation, build, testing, and required review; rerun whenever relevant code, Contract, dependencies, tests, or accepted criteria change.

Inputs

  • Component/task revision, Contract/design, Requirements, acceptance criteria, and DoD.
  • Task Workflow Level and mandatory Project Constraints.
  • Build, static check, test, review, traceability, security, and performance evidence.
  • Known failures, risks, assumptions, waivers, and deviations.

Prerequisites

  • All evidence names the exact evaluated revision and scope.
  • The implementation author cannot substitute self-assertion for required independent review.

Procedure

  1. Confirm scope, revision, DoD, Workflow Level, and mandatory checks.
  2. Reject stale or mismatched evidence.
  3. Apply level-specific baseline:
    • LIGHTWEIGHT: build/equivalent check, relevant tests, focused regression, acceptance criteria, and Quick Review.
    • STANDARD: build, static checks, unit/module tests, regression, Code Review, and acceptance criteria.
    • STRICT: build, static analysis, unit/Contract/integration/regression tests, security/performance checks as applicable, independent reviews, architecture compliance, traceability, and acceptance evidence.
  4. Add checks demanded by actual impact even if the selected level baseline is lighter.
  5. Verify every mandatory DoD item and accepted-risk authority.
  6. Record evidence, failures, residual risks, exclusions, and verdict.
  7. On PASS, mark the evaluated Component revision STABLE; on failure retain UNVERIFIED or set FAILED.

Decision Rules

  • Verdict is PASS, FAIL, BLOCKED, or USER_DECISION_REQUIRED.
  • PASS requires current evidence for every mandatory item; partial success is not PASS.
  • A waiver is a persisted risk Decision, not fabricated evidence.
  • Security-sensitive or Contract/API changes may require Strict checks in an otherwise Lightweight task.
  • Any post-review code change invalidates affected test and review evidence.

Outputs

  • Gate Record with scope, revisions, level, checks, evidence, verdict, findings, residual risks, and next action.
  • Updated Component/task state.
  • Required repair, revalidation, clarification, or integration readiness.

Failure Cases

  • Evidence references another revision or environment.
  • Required checks are unavailable or inconclusive.
  • DoD conflicts with Project Constraints.
  • Risk acceptance lacks an authorized owner.

Escalation

Use Failure Diagnosis before repeated correction. Use Clarification Gate for waivers/authority and Failure Escalation for repeated or cross-layer failure. Roll back to the responsible design/Requirement layer when evidence points upstream.

Stop Conditions

Stop after a recorded verdict. Do not integrate unless the exact Component revision is STABLE; do not continue because implementation merely exists.

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.