wiki-architect
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill identifies the repository context by executing 'git remote get-url origin' and 'git rev-parse --abbrev-ref HEAD'. These commands are used solely for metadata resolution and citation generation.
- [PROMPT_INJECTION]: The skill processes untrusted repository content which creates a surface for indirect prompt injection. 1. Ingestion points: File tree, README, and source files. 2. Boundary markers: No explicit delimiters are used to wrap ingested content. 3. Capability inventory: Metadata retrieval via git and hierarchical JSON generation. 4. Sanitization: None observed.
Audit Metadata