agentleFS
grace-methodgrace-method/grace-plugin/skills/validation/SKILL.md

validation

Skill0 starsChanged 2 months agoMITView on GitHub

Validate that implemented software achieves analysis goals. Assess goal achievement, problem resolution, and feedback loops after implementation milestones or before release.

20 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use testing.

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.
  • NVIDIASkill

    testing

    Run Python and Rust tests for physicsnemo-curator using uv + pytest with coverage reporting via pytest-cov, and cargo-nextest for Rust tests. Includes benchmark workflows with pytest-benchmark and criterion.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 20 edited versions we read keep these.

  • 6

    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:

    Use this skill after meaningful implementation milestones or before release to check whether the implemented work actually achieved the …

    line 13
  2. Distinctive:

    Use the evidence log, feature/story artifacts, test output, handoff notes, and implementation records.

    line 34
  3. Distinctive:

    If evidence is missing, say so rather than inferring success from artifact presence.

    line 36
  4. Distinctive:

    If the software shipped but the problem remains, say so explicitly.

    line 44
  5. Distinctive:

    Stop when the current path is not justified under present constraints.

    line 69
  6. Very specific:

    Write the decision dated and attributed at the moment it is made — Accept R-NNN — <name>, <YYYY-MM-DD> — in both the validation report and …

    line 73
  7. Distinctive:

    Use the templates in this skill directory when scaffolding or substantially reshaping validation artifacts.

    line 86

“Common” is measured: kept by at least 6 in 10 of the 20 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

Validation

Use this skill after meaningful implementation milestones or before release to check whether the implemented work actually achieved the goals defined earlier in the lifecycle.

Current Project Context

Analysis Goals And Problems

!cat docs/02-analysis/business-analysis.md 2>/dev/null || echo "No business analysis found."

Requirements

!cat docs/02-analysis/requirements/requirements.md 2>/dev/null || echo "No requirements found."

Implementation Status

!find docs/04-implementation/features -name "*.md" -exec head -5 {} + 2>/dev/null || echo "No features found."

Evidence Log

!cat docs/05-validation/evidence-log.md 2>/dev/null || echo "No evidence log found."

Interaction Pattern

  1. Review available evidence.
    • Use the evidence log, feature/story artifacts, test output, handoff notes, and implementation records.
    • Separate direct evidence from subjective judgment.
    • If evidence is missing, say so rather than inferring success from artifact presence.
  2. For each meaningful goal in the business analysis, assess status.
    • Mark it as met, partially met, or not met.
    • Tie each judgment to concrete evidence.
  3. For each meaningful problem, assess whether the implemented solutions actually resolve it.
    • Separate implementation completion from problem resolution.
    • If the software shipped but the problem remains, say so explicitly.
  4. Assess regression protection and test relevance.
    • Record the latest meaningful test command and result when available.
    • Note whether tests map to important criteria or only prove superficial behavior.
    • Identify regressions caught, regressions escaped, or regression risk not assessed.
  5. Assess human burden and orientation.
    • Record whether artifacts made resumption easier or harder.
    • Distinguish useful human judgment from avoidable process friction.
  6. Identify feedback items.
    • Capture new problems, new goals, missed assumptions, or opportunities discovered during validation.
  7. Route feedback items to the right stage.
    • Analysis if the problem framing or goals need revision.
    • Design if the architecture or decomposition is insufficient.
    • Implementation if the fix is mostly executional.
  8. Apply feature closeout thinking.
    • Confirm feature and story status, release notes, final test state, and retrospective readiness.
  9. Recommend the next decision.
    • Accept when goals are met well enough.
    • Iterate when specific follow-up targets are clear.
    • Stop when the current path is not justified under present constraints.
  10. Present the validation report for approval.
  • Human review decides whether to accept, iterate, or stop.
  • Write the decision dated and attributed at the moment it is made — Accept R-NNN — <name>, <YYYY-MM-DD> — in both the validation report and the evidence log. The line is the acceptance rather than a report of one, so it cannot point elsewhere for its state, and a date not written now is not recoverable later from the artifacts themselves.

Outputs

Produce or update these artifacts as needed:

  • docs/05-validation/evidence-log.md
  • docs/05-validation/validation-report.md — the acceptance decision carries its date and its decider

Use the templates in this skill directory when scaffolding or substantially reshaping validation artifacts.

Gate

Human review is required before treating the validation outcome as settled.

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.