prompt-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided draft prompts to generate optimized versions, creating a surface for indirect prompt injection where malicious instructions in the input could influence the agent's behavior.
- Ingestion points: Draft prompts provided by the user and project metadata files (e.g., package.json, go.mod, CLAUDE.md) accessed during the project detection phase.
- Boundary markers: The instructions do not specify the use of clear delimiters (like XML tags or unique markers) to isolate the untrusted user prompt from the analysis logic.
- Capability inventory: The skill is strictly limited to providing analysis and optimized text; the instructions explicitly forbid the agent from writing code, creating files, or executing commands while using this skill.
- Sanitization: There are no explicit instructions to sanitize, escape, or filter the content of the user's draft prompt before it is processed by the analysis pipeline.
Audit Metadata