agentleFS
Sign inStart free
prabhakar349prabhakar349/claude-ddd-redesign/skills/characterization-suite/SKILL.md

characterization-suite

Skill1 starsChanged 3 months agoMITView on GitHub

Defines the content and structure of a Slice's characterization test suite. Reference when assembling or reviewing slices/<id>/verification/characterization/*.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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. Very specific:

    Every Behavior in the Slice needs at least one characterization test.

    line 17
  2. Very specific:

    Every test should be traceable to the Behavior ID(s) it covers — this is what makes covers edges in the graph meaningful …

    line 23

≈ “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

Characterization Suite

Defines what the characterization suite must cover. Does not define how tests are generated or which language-specific harness conventions apply — that is Stage 6's packs/<language>/PACK.md territory (Constitution C8); this skill governs coverage obligations, which are language-independent.

Schema anchor

schemas/coverage.schema.json (slices/<id>/verification/coverage.json) is the measurable output of this suite — denominator, total, covered, passing, happy_path, and failure_path are computed directly from what the suite actually exercises. If coverage.json under-reports, the suite is incomplete, not the report wrong.

Coverage obligations (REQ-05, REQ-12, 07-stages.md Stage 6)

  • Operates at the protocol boundary — HTTP and event payloads — never at language internals (DR1). A unit test pinning an internal function is the wrong layer for this suite entirely.
  • Every Behavior in the Slice needs at least one characterization test; failure-path Behaviors are covered and reported separately from happy-path (coverage.json.failure_path) — a suite that looks complete while skipping failure paths is the characteristic silent failure this method exists to catch.
  • For async Behaviors: assert not just that an event was emitted, but which event, in what order, and with what partition key.
  • Redaction Policy verdict must be GO for every Interface this suite compares against (inventory/redaction-policy.json) — a NO-GO interface cannot be meaningfully characterized yet; that gate exists precisely to stop a suite that would compare noise.

Behavior-ID citation

Every test should be traceable to the Behavior ID(s) it covers — this is what makes covers edges in the graph meaningful (scripts/build-graph.py) and what scripts/graph-query.py coverage can check.

The harness freeze

Once this suite passes the Stage 6 gate, its hash is written to harness.lock (07-stages.md §2). The suite may not change silently during Stage 7 — any change voids the gate and requires a recorded re-approval by the QA/SRE approver.

What this skill does not cover

Adversarial gap-finding against the suite — that is agents/test-adversary.md's job, feeding back into what this suite must add before it is frozen.

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