studio-develop-golive
Warn
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: MEDIUMDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill is designed to audit environment variables and secrets within the codebase, specifically targeting configuration files such as .env. While it includes instructions to avoid leaking these secrets into chat, the automated access to sensitive file paths constitutes a potential data exposure risk.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository content as its primary input to generate documentation, creating a surface for indirect prompt injection. 1. Ingestion points: Reads the entire repository, including frameworks, dependencies, and environment configuration files (SKILL.md). 2. Boundary markers: Absent. The instructions do not define delimiters for the audited code or provide guidance to ignore instructions embedded within the codebase. 3. Capability inventory: The skill has the capability to write files to the local filesystem (docs/DEPLOY.md) and generates code prompts intended for execution in subsequent agent interactions. 4. Sanitization: Absent. There is no mention of filtering, escaping, or validating the content extracted from the repository before using it in the prompt generation process.
Audit Metadata