菜单
Reviews current code or proposed changes against stored DNA truth artifacts to identify violations, drift, and uncovered areas. Use after implementation to verify alignment with intents, contracts, and constraints, or during audits to assess truth coverage.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | reviewing-compliance |
| description | Reviews current code or proposed changes against stored DNA truth artifacts to identify violations, drift, and uncovered areas. Use after implementation to verify alignment with intents, contracts, and constraints, or during audits to assess truth coverage. |
Check whether reality matches stored truth.
Once changes are made, verify they align with the artifacts that were consulted during planning.
For each artifact that was referenced during planning:
dna get <artifact-id>
Verify the implementation satisfies the artifact. Specifically:
Search for artifacts that were NOT consulted but might be affected:
dna search "<what was changed>" --limit 10
Review any results with score > 0.5. A change to the billing service might violate a compliance artifact that was not part of the original plan.
Truth drift happens when the system has changed but the artifacts have not been updated, or when artifacts exist that no longer match reality.
Signs of drift:
When you find drift, do not silently ignore it. Either update the artifact or flag it for review.
To assess how well truth covers the system:
dna list --kind intent
dna list --kind contract
dna list --kind constraint
For each major system component, check whether:
Areas with no truth coverage are areas where agents will make unconstrained decisions. This is not always wrong (not everything needs to be codified), but high-risk areas (auth, billing, compliance, data handling) should have coverage.
During review, you may discover decisions that were made but never stored:
Capture these. See capturing-artifacts.
When truth is stale:
dna update <artifact-id> --content "<updated content>"
Include a reason for the update in the content itself, or add a label:
dna update <artifact-id> \
--content "<updated content>" \
--label updated-reason=<brief-reason>
DNA preserves version history, so the original truth is not lost.
Interactive conversation for deliberate truth changes. Acts as a truth guardian that searches DNA, surfaces conflicts, debates proposals, and only applies changes with explicit human reasoning. Use via text or voice mode in Claude Code.
Assembles DNA truth artifacts into a self-contained specification that an autonomous agent can execute against without further human guidance. Use when handing off a well-defined deliverable to an agent for extended autonomous execution.
Consults DNA truth artifacts during task planning to understand system identity before designing an implementation approach. Use at the start of any non-trivial task before choosing an architecture or decomposing work into steps. Searches for intents, contracts, and constraints that shape the solution space.
Creates escalation artifacts that define when an agent should stop autonomous work and involve a human. Use when encoding decision boundaries for agent autonomy -- the guard rails on what agents can decide alone.
Creates tradeoff artifacts that encode which concern wins when competing concerns clash. Use when two valid goals, constraints, or intents conflict and agents need to know which to prioritize.
Stores truth artifacts in DNA with correct kind, format, metadata, and semantic context. Use when codifying decisions, contracts, intents, or constraints that should persist across code changes, team transitions, and framework evolutions.