context-driven-development
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
- [NO_CODE]: The skill is entirely documentation-based, providing methodology guidelines for Context-Driven Development without any accompanying scripts or binary executables.
- [COMMAND_EXECUTION]: The 'Memory Protocol' section includes instructions to use the 'cat' command to read local files in the '.claude/context/memory/' directory. This is a low-risk, standard read operation intended for maintaining session state.
- [PROMPT_INJECTION]: Evaluation of the skill's instructions found no evidence of prompt injection patterns, safety bypass attempts, or malicious instruction overrides.
- [SAFE]: The skill defines a local documentation structure and workflow that does not involve network requests, sensitive data access, or privilege escalation.
Audit Metadata