skill-health-checker

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: When the --audit flag is used, the skill connects to https://skills.sh to retrieve security audit results and install counts for identified skills.
  • [COMMAND_EXECUTION]: The skill uses a Bash script (scripts/health-check.sh) and an embedded Python script to perform system-wide scans of agent skill directories, resolving paths and calculating file metadata.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a large ingestion surface as it reads the content of other installed skills' SKILL.md files and scripts to perform security heuristics and token counting.
  • Ingestion points: Reads files from multiple local agent directories including ~/.claude/skills/, ~/.pi/agent/skills/, and ./.agents/skills/.
  • Boundary markers: The script does not use specific delimiters or isolation techniques when processing external skill content for report generation.
  • Capability inventory: The skill has file system read access, network access (to skills.sh), and Bash execution capabilities.
  • Sanitization: The script uses regex patterns to identify threats but does not sanitize the input content before printing evidence in the consolidated report.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 06:11 AM
Security Audit — agent-trust-hub — skill-health-checker