prisma-next-feedback
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh) to automate issue creation. It also executes standard environment inspection commands such asnode -v,pnpm ls, andnode -vto gather necessary context for bug reports. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided code excerpts and command outputs to generate public artifacts. This represents an indirect prompt injection surface if malicious instructions were embedded in the processed data. However, the risk is effectively mitigated by the mandatory human-in-the-loop confirmation step before any external submission and explicit instructions to redact sensitive information.
- [DATA_EXFILTRATION]: While the skill involves sending data to external platforms (GitHub and Discord), it includes strong defensive measures: agents are instructed to aggressively redact secrets (like
DATABASE_URL), rename confidential domain details, and ensure all submissions are public-safe and explicitly approved by the user.
Audit Metadata