publish-artifacts-to-r3
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads the
r3utility from the npm registry usingnpxorbunxwhen the local executable is missing. - [EXTERNAL_DOWNLOADS]: Fetches and runs a Nix flake from the author's GitHub repository (
github:hyperlogue/r3). - [COMMAND_EXECUTION]: Executes shell commands including
npx,bunx,nix run, and ther3CLI tool to publish artifacts and access guides. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: Feedback threads and commands retrieved from the remote
r3service (SKILL.md). - Boundary markers: None specified to differentiate between reviewer comments and agent instructions.
- Capability inventory: The skill has the ability to read local files, code changes, and publish data to external infrastructure.
- Sanitization: No evidence of sanitization or validation for the incoming feedback content.
Audit Metadata