agentleFS
tuana-vntuana-vn/system-reverse-engineer/.github/skills/design-artifact-decomposition-verification/SKILL.md

design-artifact-decomposition-verification

Skill0 starsChanged 2 months agoMITView on GitHub

Independently verify design artifact decomposition against a verified TDD. Enumerate implementation-significant proposed units independently, compare them with registry ownership, detect missing/implicit/semantically unrelated ownership, and block DD progression when ownership closure is incomplete.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    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. Very specific:

    Do not trust ARTIFACT_REGISTRY_READY merely because the decomposition stage emitted it.

    line 13
  2. Distinctive:

    Create a fresh inventory without copying the registry's artifact list.

    line 26
  3. Distinctive:

    Do not infer "all units covered" from the artifact count.

    line 42
  4. Distinctive:

    If any threshold fails:

    line 76
  5. Distinctive:

    If source/TDD evidence is insufficient to determine ownership:

    line 80
  6. Distinctive:

    Only when all thresholds pass:

    line 84
  7. Distinctive:

    Use the workflow-declared output path.

    line 90
  8. Distinctive:

    Do not invent or reuse a legacy artifact path.

    line 91

≈ “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

Design Artifact Decomposition Verification

Purpose

Generator != verifier.

Do not trust ARTIFACT_REGISTRY_READY merely because the decomposition stage emitted it.

Independently re-read:

  • verified TDD
  • TDD verification
  • design artifact registry
  • authoritative requirement / verified gap where needed

Then independently enumerate implementation-significant proposed units from the TDD.

1. Independent Proposed Unit Inventory

Create a fresh inventory without copying the registry's artifact list.

A unit is implementation-significant when it creates a distinct:

  • class/interface/exception/data carrier/adapter/policy/config owner
  • responsibility or contract
  • failure behavior
  • lifecycle/dependency boundary
  • state/configuration ownership
  • verification obligation

Create:

Proposed UnitTypeTDD EvidenceWhy Implementation-Significant?

Do not infer "all units covered" from the artifact count.

2. Registry Ownership Comparison

For every independently enumerated proposed unit, require exactly one registry ownership disposition:

DEDICATED_ARTIFACT:<ART-ID>
or
SUBORDINATE_TO:<ART-ID>

Create:

Proposed UnitRegistry OwnerDedicated/SubordinateRegistry ResponsibilityExplicitly Owned?Semantic Match?Verdict

Rules:

  • a JavaDoc mention, import, method signature, throws, @see, dependency, or nearby prose is not ownership
  • an artifact cannot own a semantically unrelated unit merely because it references it
  • self-referential labels such as SUBORDINATE_TO:ART-005 for the same ART-005 are invalid ownership semantics unless the registry clearly identifies a separate subordinate proposed unit
  • existing modified components are not "subordinate to themselves"; their artifact ownership type should describe the artifact itself, while subordinate proposed units must be separately named

3. Required Quantitative Gates

independent proposed-unit inventory completed = YES
proposed-unit ownership coverage = 100%
semantic owner match = 100%
unowned implementation-significant units = 0
implicitly-owned units = 0

If any threshold fails:

ARTIFACT_DECOMPOSITION_VERIFICATION_FAILED

If source/TDD evidence is insufficient to determine ownership:

ARTIFACT_DECOMPOSITION_VERIFICATION_BLOCKED

Only when all thresholds pass:

ARTIFACT_DECOMPOSITION_VERIFIED

4. Output

Use the workflow-declared output path. Do not invent or reuse a legacy artifact path.

<!-- workflow-status-contract:start -->
emitted_statuses:
  - ARTIFACT_DECOMPOSITION_VERIFIED
  - ARTIFACT_DECOMPOSITION_VERIFICATION_FAILED
  - ARTIFACT_DECOMPOSITION_VERIFICATION_BLOCKED
<!-- workflow-status-contract:end -->

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.