init-agents-md
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted data from repository files to generate a new markdown file.
- Ingestion points: Processes content from
README.md,pyproject.toml,package.json, and various configuration files. - Boundary markers: The instructions do not specify any delimiters or safety markers to differentiate between legitimate project metadata and potentially malicious instructions embedded in the source files.
- Capability inventory: The skill utilizes file read and write capabilities within the project root.
- Sanitization: No sanitization or validation of the ingested content is defined before it is formatted into the final
AGENTS.mdfile. - [NO_CODE]: The skill consists entirely of markdown-based instructions and does not include any executable scripts or binary files, which limits the scope of potential technical attacks to prompt-based vectors.
Audit Metadata