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

stage-gate

Skill1 starsChanged 53 days agoMITView on GitHub

Evaluate whether a major lifecycle Stage has complete valid Artifacts, passed required validation and Definition of Done, and reached the configured Target Stage. Use at each relevant Stage boundary before continuing or pausing.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run when target-relevant work for a Stage appears complete, before entering downstream work, or when reassessing a paused/failed Stage.

    line 14
  2. Very specific:

    Check whether all target-required Stage Artifacts are complete and sufficiently VALID.

    line 30
  3. Distinctive:

    Check Artifact Validation and all applicable lower-level Gate verdicts.

    line 31
  4. Distinctive:

    Check every Stage DoD item at the effective Workflow Level.

    line 32
  5. Distinctive:

    Check traceability, accepted risks, unresolved deviations, assumptions, and Clarifications.

    line 33
  6. Distinctive:

    Do not enter a later stage because it exists in the lifecycle.

    line 47
  7. Distinctive:

    Use Clarification Gate for intent/acceptance, Failure Diagnosis for failed evidence, Change Impact for drift, and Workflow Assessment when …

    line 64
  8. Very specific:

    Stop after the recorded verdict.

    line 68

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Stage Gate

Purpose

Control progression by evidence and stop precisely at the user's Target Stage.

Trigger

Run when target-relevant work for a Stage appears complete, before entering downstream work, or when reassessing a paused/failed Stage.

Inputs

  • Current Stage, Target Stage, Stage scope, Workflow Level, and Stage DoD.
  • Required Artifact IDs/revisions/statuses and validation evidence.
  • Applicable Clarification, Artifact, Component Quality, and Integration Gate results.
  • Traceability, open questions, assumptions, decisions, risks, deviations, and Current Minimum Necessary Path.

Prerequisites

  • Normalize the stage/target and identify only Artifacts required for the declared scope.
  • Gate evidence must apply to current revisions.

Procedure

  1. Check whether all target-required Stage Artifacts are complete and sufficiently VALID.
  2. Check Artifact Validation and all applicable lower-level Gate verdicts.
  3. Check every Stage DoD item at the effective Workflow Level.
  4. Check traceability, accepted risks, unresolved deviations, assumptions, and Clarifications.
  5. Determine whether failures require rework, clarification, or rollback to a responsible earlier layer.
  6. Determine whether Current Stage equals Target Stage.
  7. Record PASS, FAIL, BLOCKED, or USER_DECISION_REQUIRED with evidence and next action.
  8. On PASS below target, update State and expose only the next ready Minimum Path node.
  9. On PASS at target, create a Checkpoint, set project status PAUSED_AT_STAGE, report outputs/residual risks, and stop.

Decision Rules

  • Artifact existence is not completeness or validation.
  • Missing/currently stale required evidence prevents PASS.
  • Waived criteria require persisted authority and risk acceptance.
  • Rework only affected scope and recompute the Minimum Necessary Path.
  • Reaching Target Stage means PAUSED_AT_STAGE, never PROJECT_COMPLETED.
  • Do not enter a later stage because it exists in the lifecycle.

Outputs

  • Stage Gate Record with Stage/target, revisions, DoD checks, evidence, verdict, findings, and next action.
  • Updated Current Stage/project status and path state.
  • Target delivery summary and Checkpoint when paused.

Failure Cases

  • Stage DoD or target is ambiguous.
  • Evidence is stale, conflicting, or absent.
  • Required user Acceptance has not occurred.
  • State and repository revisions disagree.

Escalation

Use Clarification Gate for intent/acceptance, Failure Diagnosis for failed evidence, Change Impact for drift, and Workflow Assessment when Gate depth is insufficient. Roll back only to the responsible layer.

Stop Conditions

Stop after the recorded verdict. On target PASS, stop all downstream work at PAUSED_AT_STAGE; on failure/block, expose only bounded corrective action.

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.