investigation-open

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • Local Configuration Access: In SKILL.md, the skill is configured to load a specific file from ~/.claude/plugins/config/claude-for-legal/employment-legal/CLAUDE.md. This path is consistent with the tool's defined environment for legal investigation management.
  • Modular Skill Integration: The instructions utilize an internal-investigation reference skill to handle substantive intake and templating, following a standard modular pattern for agent task delegation.
  • Data Ingestion Surface: The skill processes user-provided allegation descriptions to generate investigation documentation in SKILL.md.
  • Ingestion points: allegation description passed via the argument-hint field in SKILL.md.
  • Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are present in the intake workflow.
  • Capability inventory: File access to plugin configurations and invocation of reference skills in SKILL.md; creation of persistent investigation logs.
  • Sanitization: The input is used directly to populate investigation checklists and templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 07:01 PM
Security Audit — agent-trust-hub — investigation-open