forge-prompt
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a prompt generator that takes untrusted user input and reformats it into instructions for a downstream agent session.
- Ingestion points: The skill accepts a 'short, unclear ask' from the user during the initial interaction (SKILL.md).
- Boundary markers: The skill structures its output using specific headers ('What this touches', 'Off limits', 'Must keep working', 'Done when'), though the user's raw input is included as the primary subject.
- Capability inventory: The skill is governed by a '' forbidding it from writing code or editing project files. It utilizes repository-reading capabilities to inform the interview process.
- Sanitization: The instructions do not specify any validation or sanitization steps for the user-provided input before inclusion in the final output block.
Audit Metadata