Skills
Navigate
skills/rheinmir/setup/health-check/Gen Agent Trust Hub

health-check

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses python3 to execute local scripts located in the harness/scripts/ directory, specifically health-check.py and okf-check.py, to perform diagnostic tasks and repository updates.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources and local configuration files, which presents a surface for indirect prompt injection.
  • Ingestion points: The health-check.py script fetches version.json and file fingerprints from a remote branch (e.g., orca) in the Rheinmir/setup repository.
  • Boundary markers: None explicitly defined in the skill instructions to separate remote data from the agent's internal logic.
  • Capability inventory: The skill has the capability to execute shell commands (python3, git checkout) and modify local files via the --update and --migrate flags.
  • Sanitization: There is no evidence of sanitization or validation of the remote content before it is processed by the local scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 03:25 AM