is-this-a-problem
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill accesses configuration files and matter-specific data located in the
~/.claude/plugins/config/claude-for-legal/directory. This is standard behavior for the tool to retrieve its logic and manage case-specific context. No access to general system secrets or sensitive user directories was detected. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from user questions and processes contents from
matter.md. Ingestion points: User input[the question]and external filesCLAUDE.mdandmatter.md. Boundary markers: Absent. Capability inventory: File read and write operations restricted to the plugin's internal directory structure. Sanitization: No explicit sanitization or filtering of input data before processing. The risk is considered minimal as the skill performs pattern-matching for text-based triage without executing commands or making network requests.
Audit Metadata