agentleFS
Sign inStart free
agent-ixagent-ix/spec-skills/spec-evidence-analysis/SKILL.md

spec-evidence-analysis

Skill1 starsChanged 6 months agoMITView on GitHub

Define verification methods and evidence artifacts for requirements.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Make sure every requirement has a defined verification method and a concrete artifact that proves it was verified.

    line 8
  2. Very specific:

    After requirements are written but before spec-matrix builds the test matrix.

    line 12
  3. Distinctive:

    Rule: Every implementation task MUST cite at least one evidence artifact.

    line 36
  4. Very specific:

    Resolve "test" defaults: if an FR is tagged test, the matching test case must already exist or be queued in spec-matrix.

    line 41
  5. Very specific:

    Every FR/StR/NFR has both verification_method and at least one evidence entry.

    line 76

≈ “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

Verification & Evidence Strategy

Make sure every requirement has a defined verification method and a concrete artifact that proves it was verified.

When to use

  • After requirements are written but before spec-matrix builds the test matrix.
  • During spec-review when "how do we know this is satisfied?" is unclear.
  • Whenever a new NFR is added (NFRs are most prone to missing evidence).

Verification methods (ISO/IEC/IEEE 29148)

MethodUse when
testThe requirement is exercisable by automated or manual test execution. Default.
analysisThe requirement is proven by calculation, modeling, or static reasoning (e.g. complexity bound, cryptographic argument).
inspectionThe requirement is satisfied by visual or structural review of the artifact (e.g. license headers, doc presence).
demonstrationThe requirement is satisfied by observing operation in a realistic environment without instrumentation (e.g. UX flow walkthrough).

Evidence artifacts

ArtifactPairs with
test_casetest — points to a TC-XXX file or test path
analysis_reportanalysis — points to a doc/notebook
inspection_checklistinspection — points to a checklist file
demo_recordingdemonstration — points to a recording or scripted walkthrough
metricany — observability metric that proves runtime behavior
logany — log event proving the path executed

Rule: Every implementation task MUST cite at least one evidence artifact. A task with no evidence is unfinishable.

Process

  1. Tag every FR/StR/NFR with verification_method and evidence fields in its frontmatter.
  2. Resolve "test" defaults: if an FR is tagged test, the matching test case must already exist or be queued in spec-matrix.
  3. Resolve non-test methods: for analysis/inspection/demonstration, name the artifact path explicitly. No "TBD".

Deliverable template

Per-requirement frontmatter (add to existing FR/StR/NFR files):

verification_method: test          # test | analysis | inspection | demonstration
evidence:
  - kind: test_case
    ref: tests/integration/test_login.py::test_invalid_credentials
  - kind: metric
    ref: auth_failure_total

Summary deliverable in spec/analysis/evidence.md:

# Verification & Evidence Summary

| Req | Method | Artifact |
|-----|--------|----------|
| FR-003 | test | tests/integration/test_login.py::test_login_happy_path |
| FR-010 | analysis | docs/analysis/storage-bound.md |
| NFR-002 | demonstration | docs/demos/uptime-walkthrough.md |
| NFR-005 | test + metric | tests/perf/test_p99_latency.py, metric `http_request_duration_seconds` |

## Open gaps

None.

Acceptance Criteria

  • Every FR/StR/NFR has both verification_method and at least one evidence entry.
  • No evidence: TBD, no missing refs.
  • The summary table has zero open gaps before spec-to-plan is invoked.

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.

Check your setup, free