unblocked-context-search-code
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation instructs the agent to verify the existence of the 'unblocked' CLI tool using 'command -v' and to execute searches via the 'unblocked context-search-code' command. These are legitimate operations for using the vendor's provided search utility.
- [EXTERNAL_DOWNLOADS]: The skill's primary function is to search and retrieve code from external version control platforms, including GitHub, GitLab, Bitbucket, and Azure DevOps. This interaction with well-known services is the intended and documented behavior of the tool.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it processes code content from external repositories. However, this risk is mitigated by the lack of dangerous capabilities that could be exploited through the ingested data. 1. Ingestion points: Output from the 'context_search_code' command. 2. Boundary markers: Not explicitly defined in the prompt instructions. 3. Capability inventory: Access to the 'unblocked' CLI and standard agent tools for reading or searching files. 4. Sanitization: No specific filtering or escaping is applied to the retrieved code results.
Audit Metadata