teach-impeccable

Pass

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8).
  • Ingestion points: The skill reads external codebase content including README, documentation, package.json, configuration files, components, and style guides.
  • Boundary markers: There are no explicit instructions to the agent to ignore or escape embedded instructions found within the scanned codebase files.
  • Capability inventory: The agent has the capability to read files, prompt the user via AskUserQuestionTool, and write to the filesystem (specifically CLAUDE.md).
  • Sanitization: The skill does not define any sanitization or validation logic for the content gathered from the codebase before it is synthesized and written to the persistent CLAUDE.md file.
  • Risk: If a scanned file contains malicious instructions (e.g., hidden in a README or code comment), the agent might incorporate those instructions into the CLAUDE.md file, which typically influences the agent's behavior in all future sessions within that project.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 8, 2026, 12:20 AM
Security Audit — agent-trust-hub — teach-impeccable