Skip to content
Back to skills

Senmu Build Assurance

ASecurity

Conduct explicit audits, decision POCs, comparisons or disputed-cause reviews with scoped evidence. Audit scope does not imply reviewer independence. Not for routine domain self-checks or production implementation.

  • 2 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added September 2, 2026
ai-agentsgogit

Security analysis

A100/100

Pro scans all 13 files and shows the line behind each finding

Scanned September 28, 2026

npx -y skills add SenMuShare/senmu-buildos --skill senmu-build-assurance --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Senmu Build Assurance?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Senmu Build Assurance
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/senmushare-senmu-build-assurance/badge)](https://www.skillsdirectory.com/skills/senmushare-senmu-build-assurance)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: senmu-build-assurance
description: "Conduct explicit audits, decision POCs, comparisons or disputed-cause reviews with scoped evidence. Audit scope does not imply reviewer independence. Not for routine domain self-checks or production implementation."
---

# Governance Assurance

Reviews are read-only by default; an authorized POC may write its isolated experiment materials, not the audited product or production state. Freeze the subject, version, scope, and standard; distinguish facts, inferences, and unknowns with reviewable evidence. A verdict does not itself authorize remediation.

## Route by Outcome

- For a decision POC, blind test, controlled experiment, ledger, or reproduction, read [Reproducible POC Governance](references/reproducible-poc-governance.md).
- For a code, architecture, governance, delivery, or whole-project review, read [Independent Review and Evidence Grading](references/independent-review-and-evidence-grading.md). Use `exhaustive_source` only when the user explicitly requests every file, function, or existing comment.
- For resumable or explicitly coverage-tracked Git reviews, use [Frozen Review Execution](references/frozen-review-execution.md). The helper is optional and does not change reviewer identity or domain ownership.

Explicit audits, cross-domain disputes and required independent verdicts use Assurance. Routine consistency checks stay with the domain; G3-G4 alone does not activate it. Choose reviewer identity separately from audit scope. Read only applicable Engineering guidance.

## Core Contract

- Declare the review as `independent`, `peer`, or `evidence-based self-review`; do not claim independence without demonstrable separation.
- Record the frozen target, coverage, evidence source and freshness, excluded scope, and stopping conditions.
- Evidence supports only what it observes. Static analysis, tests, production facts, and independent review are not interchangeable. Execution completion does not establish quality or approval.
- Seek counterevidence before assigning status, P0-P3, impact, minimum remediation, and re-review conditions.
- Keep `not_assessed`, `inconclusive`, `resolved_unverified`, and `verified_resolved` distinct.
- Review authority does not permit modification, release, deletion, or production changes. Return remediation to its domain owner. When the same request already authorizes repair, continue there after findings without another generic approval; review-only requests remain read-only.

Use the project's durable task owner for multi-stage reviews. Handoffs carry findings, evidence, scope, target outcomes, and re-review conditions, not copied standards.

Files in this skill

  • SKILL.md1.8 KB
  • agents/openai.yaml279 B
  • assets/poc-experiment-governance/EXPERIMENT_REGISTER.template.md530 B
  • assets/poc-experiment-governance/experiment-package/DECISION.template.md275 B
  • assets/poc-experiment-governance/experiment-package/EXPERIMENT.template.md693 B
  • assets/poc-experiment-governance/experiment-package/PLAN.template.md237 B
  • assets/poc-experiment-governance/experiment-package/RESULTS.template.md357 B
  • assets/poc-experiment-governance/experiment-package/experiment-manifest.template.json210 B
  • assets/review-governance/ASSURANCE_REVIEW_REPORT.template.md2 KB
  • assets/review-governance/EXHAUSTIVE_SOURCE_REVIEW_CONTROL.template.json629 B
  • references/POC可复现实验治理规范.md16.4 KB
  • references/独立审查与证据分级规范.md9.8 KB
  • scripts/validate_exhaustive_source_review.py20.6 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.