code-standards

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to read and "obey" project-specific files such as tasks/lessons.md, README, and CONTRIBUTING during its orientation phase. This creates a surface for indirect prompt injection, as malicious instructions embedded within a repository's documentation could be used to override the agent's intended behavior.
  • Ingestion points: The agent is directed to read tasks/lessons.md (Step 1) and subsystem-specific documentation files.
  • Boundary markers: The instructions lack boundary markers or warnings to treat external project content as untrusted data.
  • Capability inventory: The workflow grants the agent broad file system access to modify code, run test suites, and update documentation.
  • Sanitization: There is no evidence of sanitization or filtering applied to the content read from the repository files.
  • [EXTERNAL_DOWNLOADS]: The README.md file contains instructions for the user to install the skill using npx @tamng0905/builder-essential-skills. This references a scoped Node.js package associated with the author's infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 03:28 PM
Security Audit — agent-trust-hub — code-standards