rlm-audit

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation provides an example command to run a local script scripts/audit_cache.py. This is intended functionality for the described task and uses local execution without elevated privileges.
  • [DATA_EXPOSURE]: The skill reads from the local filesystem to compare files against a manifest. This operation is restricted to checking for file existence and metadata to identify gaps in the cache, which aligns with its stated purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes project documentation and manifests from the local filesystem. As it ingests external data to perform its audit, it possesses a standard attack surface for indirect prompt injection, though no specific exploits are present in the static instructions.
  • Ingestion points: Reads project manifest and local directory structure.
  • Boundary markers: None explicitly defined in the instructions.
  • Capability inventory: Executes local Python script and writes to local text and CSV files.
  • Sanitization: None explicitly defined for file names or manifest entries.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 08:07 PM
Security Audit — agent-trust-hub — rlm-audit