tighten
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided documents, which could contain malicious instructions designed to bypass agent constraints or influence behavior during the compression process.\n
- Ingestion points: The document specified in the
path-to-docargument is read into the agent context as part of the diagnosis and indexing steps.\n - Boundary markers: Absent; the instructions do not define specific delimiters or provide guidance to the agent to treat the document content as untrusted data.\n
- Capability inventory: The skill is granted access to powerful tools including
Read,Write,Edit,Bash,Grep, andGlob, which could be exploited if an injection is successful.\n - Sanitization: Absent; the agent reads and indexes facts from the document directly without filtering or validation of the input content.
Audit Metadata