# Audit · LangGuard-AI

> Use this skill ONLY when the user explicitly invokes /scope-mcp:audit (or asks for a "compliance audit"). Runs a compliance audit against the curated action-surface database. Accepts MCP tool ids, API actions, connector wildcards (salesforce.), bare platform names, or a prose description of the agent. Returns a deterministic risk + regulatory exposure report (SOX, GDPR, PCI, HIPAA) with concrete recommendations. Handles both design-time scoping ("should I build this?") and run-time pre-flight ("about to execute these tools — any issues?") — choose the framing in Step 3 based on what the user is actually doing.

## Facts
- Page: https://tashan.sh/capability/skill-langguard-ai-audit
- tashan id: skill:LangGuard-AI/audit
- Source: https://github.com/LangGuard-AI/scope-mcp
- Type: skill
- Category: security
- tashan score: 50.0 / 100
- Adoption: 22.0
- Upkeep: 78.0
- Freshness: 92.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- License: Apache-2.0
- Official: no

## Install

```sh
cp -r audit ~/.claude/skills/
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-08-14 by tashan (https://tashan.sh) from public evidence. Scorer s5.