instrument-integration

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions (Step 7) create an indirect prompt injection surface by directing the agent to run linter or formatting scripts from the user's project configuration.
  • Ingestion points: Project configuration files such as package.json (as described in SKILL.md).
  • Boundary markers: Absent; there are no instructions to verify the content of the scripts before execution.
  • Capability inventory: Shell command execution via package managers (e.g., npm run lint).
  • Sanitization: Absent.
  • [EXTERNAL_DOWNLOADS]: The skill integrates official PostHog SDKs by downloading them from public package registries (NPM, PyPI, etc.) and referencing official vendor repositories on GitHub.
  • [CREDENTIALS_UNSAFE]: A hardcoded PostHog project token (phc_jE9kXU0depRekiuabVROlxxkIXn95NqsNO3qB4qNKtl) was identified in the configuration of an example project within the references (BurritoConsiderationClient.xcscheme). This is a vendor-provided demo credential for the included sample applications.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 02:15 AM
Security Audit — agent-trust-hub — instrument-integration