codebase-research
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed as a focused, read-only utility for codebase research. It contains explicit instructions forbidding file modifications or any operations that mutate the state of the branch.
- [COMMAND_EXECUTION]: The skill uses the dxs CLI (a vendor-specific tool) and jq for inspecting code and processing JSON envelopes. These commands are restricted to read-only exploration (e.g., 'dxs source explore', 'dxs configuration get') and are appropriate for the skill's stated purpose.
- [PROMPT_INJECTION]: The skill handles an indirect prompt injection surface by ingesting and analyzing untrusted TypeScript code from Datex Studio flows. This risk is inherent to the research task and is mitigated by the skill's read-only constraints and its requirement for a structured 'Answer / Sources / Caveats' response format, which ensures auditability of the sources used.
Audit Metadata