to-prd
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell scripts (
scripts/next-prd-index.shandscripts/publish-prd.sh) to determine the next file index and to interact with the GitHub CLI (gh). These scripts are bundled with the skill and utilize standard shell quoting to prevent argument injection. - [DATA_EXFILTRATION]: The skill publishes the generated PRD content to GitHub Issues via the
ghCLI tool. This action is the primary intended purpose of the skill and targets the repository currently in use by the agent, representing legitimate tool usage. - [PROMPT_INJECTION]: The skill ingests potentially untrusted data from the conversation context and codebase to generate document content. While this creates a surface for indirect prompt injection, the skill's capabilities are limited to writing local files and creating tracker issues. The shell scripts safely handle input variables using double-quoting, mitigating the risk of command injection from processed data.
- [EXTERNAL_DOWNLOADS]: No unauthorized external downloads or remote code executions were found. The skill relies on pre-installed system utilities and local scripts.
Audit Metadata