codebase-knowledge-builder
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: No direct prompt injection or safety bypass patterns were detected within the skill's instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data from local code repositories, which represents a potential injection surface.
- Ingestion points: The skill reads source code, configuration files (e.g., package.json, Dockerfile), and documentation across the entire repository.
- Boundary markers: Absent. There are no explicit instructions or delimiters used to separate repository content from the agent's internal instructions.
- Capability inventory: The skill utilizes shell execution (via
analyze.sh) and broad file-reading capabilities. - Sanitization: Absent. The skill does not specify any filtering or sanitization of the content found within the analyzed files before processing.
- [COMMAND_EXECUTION]: The skill executes a local shell script (
scripts/analyze.sh) to gather repository metrics. The script follows security best practices by resolving paths to absolute locations and using quoted variables to prevent common command injection vulnerabilities.
Audit Metadata