diagnosing-mcp
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions facilitate a surface where the agent ingests and modifies untrusted local data.
- Ingestion points: The agent is directed to read from several local configuration files, including
~/.zcode/cli/config.json,.agents/mcp.json, and repository-specific.zcode/config.jsonfiles. - Boundary markers: The instructions do not define delimiters or specify that the agent should ignore instructions embedded within the configuration files it reads.
- Capability inventory: The skill assumes and encourages the use of file-read and file-write capabilities across the local filesystem.
- Sanitization: There is no mention of sanitizing or validating the contents of the JSON configuration files before the agent processes or applies them.
Audit Metadata