agency-status
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands like 'ls', 'find', and 'test' to discover files and check for the presence of specific tool suites in the user's home directory ('~/.claude/skills/').- [INDIRECT_PROMPT_INJECTION]: The skill reads and processes data from audit reports and proposal files to extract scores and pricing metrics.
- Ingestion points: Reads content from files such as 'AGENCY-ONBOARD-.md' and 'AGENCY-PROPOSAL-.md'.
- Boundary markers: None present.
- Capability inventory: Directory listing and file existence checks via 'bash'.
- Sanitization: None; extraction relies on simple string or regex matching.
Audit Metadata