skills/beclab/olares/olares-files/Gen Agent Trust Hub

olares-files

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to list and read the contents of files, creating an attack surface where untrusted data could influence agent behavior. \n
  • Ingestion points: Data is ingested via files ls, files cat, files archive entries, and files edit operations. \n
  • Boundary markers: The instructions do not define delimiters to separate file content from the agent's instructions. \n
  • Capability inventory: The skill provides access to powerful tools for file modification, deletion, network share mounting (SMB/NFS), and cloud storage transfers. \n
  • Sanitization: The underlying CLI tool enforces path-traversal protection by blacklisting '.' and '..' segments in paths. \n- [COMMAND_EXECUTION]: The skill serves as a wrapper for the olares-cli utility. While it provides administrative capabilities such as changing file ownership to root, it includes mandatory safety steps requiring the agent to confirm user intent and use secure inputs for credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:14 AM
Security Audit — agent-trust-hub — olares-files