buildit
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project environment files to detect the development stack, which could be exploited if those files contain malicious instructions.
- Ingestion points: The agent reads local configuration and source files such as
package.json,pyproject.toml,go.mod, and framework markers likemanage.pyornext.config.*(Step 0). - Boundary markers: No delimiters or instructions are provided to the agent to disregard potentially malicious content embedded within these read files.
- Capability inventory: The skill allows the agent to generate and implement code for UI, API, and database layers, and encourages running verification tools via the
/gatecommand. - Sanitization: The skill does not include steps to validate or sanitize the contents of the detected configuration files before using them to inform its development logic.
Audit Metadata