workspace-search
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use external CLI search tools such as tgrep, ast-grep, ripgrep (rg), and grep. It also provides guidance on starting, monitoring, and stopping a background indexing service using tgrep serve.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted workspace data when searching for code or text patterns. 1. Ingestion points: Workspace files processed by the search tools. 2. Boundary markers (absent): No delimiters are provided to the agent to ignore embedded instructions in searched files. 3. Capability inventory: Execution of CLI search tools and service management. 4. Sanitization (absent): No validation of file content is specified.
Audit Metadata