context-maintain
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process existing project knowledge to improve coherence and organization, which creates an attack surface for indirect prompt injection if the sources contain malicious instructions.
- Ingestion points: Step 1 and Step 2 involve reading and comparing sources across project directories to identify dependencies and problems.
- Boundary markers: The instructions do not define explicit delimiters or headers to differentiate ingested project data from the agent's internal reasoning or system instructions.
- Capability inventory: The skill specifies capabilities for reading existing project guidance and applying changes to "moved or replaced content" and "navigation," implying file-system write access.
- Sanitization: There are no instructions provided for sanitizing or escaping the content of the processed files before they are analyzed or rewritten.
Audit Metadata