gsp-project-review
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute
git diffandgh pr list. These commands are used to identify code changes and retrieve repository metadata, which are standard operations for a QA review workflow. - [PROMPT_INJECTION]: The skill ingests untrusted data from the codebase, build logs, and design specs, creating an indirect prompt injection surface. * Ingestion points: Reads source files listed in build logs and the output of git diff commands. * Boundary markers: No explicit delimiters or boundary instructions are implemented to isolate untrusted codebase content from the agent instructions. * Capability inventory: File system access (Read/Write), Bash execution (git/gh), and sub-agent spawning. * Sanitization: No filtering or sanitization of the ingested codebase content is performed.
Audit Metadata