error-diagnosis
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFE
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill is designed to ingest and process untrusted data in the form of user-provided error messages and stack traces. Ingestion points: Step 1 and 2 of the Diagnosis Workflow in SKILL.md. Boundary markers: Absent; the skill does not define specific delimiters for the error content. Capability inventory: The skill references MCP tools for memory retrieval (memory_sessions, reasoning_query) but contains no subprocess calls, file-write operations, or network requests in its own definition. Sanitization: Absent.
- [Privilege Escalation] (SAFE): The text mentions commands like 'sudo', 'chmod', and 'chown' as suggested fixes for permission errors. These are instructional references only and are not executed automatically.
- [External Downloads] (SAFE): The text suggests 'npm install' and 'pip install' as manual steps for resolving missing dependencies. No automated package installation or remote script execution is present.
Audit Metadata