prompt-optimizer

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes external, untrusted content.
  • Ingestion points: Draft prompts provided by the user and local project documentation files (specifically CLAUDE.md) are read during the project detection phase.
  • Boundary markers: The instructions include multiple explicit directives to maintain an 'Advisory role only' and forbid implementing code or executing commands (e.g., 'Do NOT write code, create files, run commands, or take any implementation action').
  • Capability inventory: The skill is restricted to generating text analysis and optimized prompt templates; it contains no internal logic for file modification, network communication, or shell command execution.
  • Sanitization: No explicit sanitization or escaping of user input is implemented; the skill relies on restrictive behavioral instructions to mitigate risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 07:42 AM
Security Audit — agent-trust-hub — prompt-optimizer