cs-build-create-prd
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
codespringCLI andnodecommands to retrieve project status and modify mindmap nodes. It specifically reads from.codespring/config.jsonto identify the project scope. - [EXTERNAL_DOWNLOADS]: The
allowed-toolssection includesnpx @codespring-app/cli, which triggers the download of the CodeSpring CLI package from the NPM registry if not present locally. This is a recognized vendor resource. - [REMOTE_CODE_EXECUTION]: The execution of
npxwith a remote package allows for the running of externally sourced code within the shell environment. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it is instructed to "deep-dive the real code" to extract architectural details and shared contracts.
- Ingestion points: Local source code files and directories processed during the feature analysis phase.
- Boundary markers: The instructions lack specific delimiters or warnings to the agent regarding the separation of data (code) from instructions.
- Capability inventory: The agent has capabilities to read files, execute shell commands, and perform network POST requests to the CodeSpring API.
- Sanitization: There are no instructions for sanitizing or escaping the content read from the codebase before it is interpolated into the context for PRD generation.
Audit Metadata