Skip to content
Back to skills

Do Integration Audit

ASecurity

Trace whether a feature is reachable and correctly wired across implementation, entrypoints, configuration, tests, and docs.

  • 26 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 23, 2026
ai-agentsdocumentation

Works with

  • cli

Security analysis

A100/100

Scanned September 23, 2026

npx -y skills add tomcounsell/ai --skill do-integration-audit --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Do Integration Audit?

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

Security grade badge for Do Integration Audit
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tomcounsell-do-integration-audit-ai/badge)](https://www.skillsdirectory.com/skills/tomcounsell-do-integration-audit-ai)

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: do-integration-audit
description: "Trace whether a feature is reachable and correctly wired across implementation, entrypoints, configuration, tests, and docs."
---

# Do Integration Audit

Search for the feature and synonyms in source, imports, routes, CLI commands, events, config, migrations, tests, and docs. Build an integration map of implementation, entrypoints, tests, documentation, configuration, and migrations.
Check orphan code, dead entrypoints, partial wiring, missing integration coverage, undocumented entrypoints, configuration gaps, stale references, inconsistent interfaces, non-reusable interfaces, internal naming drift, external naming drift, and missing error boundaries.
Before reporting, re-read every cited location. A claim that nothing handles a case requires project-wide search, including watchdogs, sibling services, hooks, and migration paths. Runtime claims require tracing actual bodies, guards, early returns, idempotency, flags, and exception paths; function-call proximity is not proof of execution. Mark unproven behavior as suspected.
Group verified findings by consequence and offer specific fixes. For a severity filter, omit lower-severity results but preserve incomplete-check disclosures. Audit-only requests produce findings; audit-and-fix requests continue through relevant repairs and verification.

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.