prospector
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes subagents to execute shell-based Git commands like
git log,git blame, andgit showfor historical analysis of the repository. - [PROMPT_INJECTION]: The skill ingests untrusted source code and dependency manifests, creating an indirect prompt injection surface.
- Ingestion points: Codebase files (e.g., package.json, Cargo.toml, source files) read during various analysis phases.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the analysis or design prompt templates.
- Capability inventory: Execution of git commands, file writing to documentation and project memory, and dispatching other specialized agents.
- Sanitization: No sanitization of ingested code content is performed before interpolation into subagent prompts.
Audit Metadata