skills/av/facts/facts-refine/Gen Agent Trust Hub

facts-refine

Pass

Audited by Gen Agent Trust Hub on May 17, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill interacts with a local CLI tool named facts and its aliases (ll, at, rt, rm, ls) to manage project documentation. This is the primary intended function of the skill.
  • [COMMAND_EXECUTION]: The skill can generate and propose shell commands for fact validation (e.g., using grep). The instructions enforce a human-in-the-loop pattern, requiring the agent to present findings and wait for user approval before applying any changes, which effectively mitigates risks associated with automated command execution.
  • [DATA_EXPOSURE]: The skill reads the project's fact sheet to identify draft items. There is no evidence of unauthorized sensitive file access or network-based exfiltration. The operations are confined to the project's own management data.
Audit Metadata
Risk Level
SAFE
Analyzed
May 17, 2026, 07:37 PM
Security Audit — agent-trust-hub — facts-refine