policy-drafting

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a legal drafting tool, reading jurisdictional and matter-specific context from local configuration files within the plugin configuration directory. No malicious code, data exfiltration attempts, or unauthorized command executions were identified.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading configuration (CLAUDE.md) and matter-specific files (matter.md) to inform its drafting process. This is a common pattern for such tools and is considered safe in this local context.
  • Ingestion points: SKILL.md (references ~/.claude/plugins/config/claude-for-legal/employment-legal/CLAUDE.md and matter.md).
  • Boundary markers: Absent. The skill does not implement delimiters or instructions to ignore potential commands embedded in the ingested files.
  • Capability inventory: SKILL.md (drafts policy text and writes to the local directory ~/.claude/plugins/config/claude-for-legal/employment-legal/matters/).
  • Sanitization: Absent. No validation or filtering is applied to the ingested content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 03:58 PM
Security Audit — agent-trust-hub — policy-drafting