health-check

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill runs standard quality gate commands (e.g., npx tsc, npx vitest, npx eslint) which are hardcoded in the skill's scripts. These are benign diagnostic commands.\n- [EXTERNAL_DOWNLOADS]: Utilizes npx to execute tools, which may download packages from the npm registry. As npm is a well-known and trusted service, this is considered safe behavior.\n- [PROMPT_INJECTION]: The skill processes untrusted repository content through diagnostic tools, creating a surface for indirect prompt injection.\n
  • Ingestion points: Tool outputs in scripts/run-health-check.sh read the codebase.\n
  • Boundary markers: SKILL.md contains explicit 'NEVER' instructions to stop the agent from analyzing or reacting to the tool output.\n
  • Capability inventory: Limited to diagnostic shell commands with no file-write or external network access.\n
  • Sanitization: The bash script sanitizes output by truncating text and removing control characters before returning a summary to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 10:05 PM