create-instructions
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is subject to indirect prompt injection because it analyzes untrusted source code. Maliciously crafted content in the project's code could be designed to influence the instructions the agent generates in the
AGENTS.mdfile. \n - Ingestion points: Source files read during the analysis phase. \n
- Boundary markers: None present. \n
- Capability inventory: File system write access to the project root for creating
AGENTS.md. \n - Sanitization: No content validation or filtering is performed on the data read from the files.
Audit Metadata