nest-boot-maintainer
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied evidence and code reproductions which are then used to run local tests and generate GitHub issues. While this creates a potential surface for instructions embedded in bug reports to influence agent behavior, the skill includes robust mitigations.
- Ingestion points: Minimal reproduction code, configuration files, and environment details provided in the user prompt or local files.
- Capability inventory: Shell access for running tests (
pnpm,nx,jest), network access via the GitHub CLI (gh), and repository management tools (git). - Boundary markers: The skill provides structured templates for issues and pull requests to help contain external data.
- Sanitization: Instructions explicitly mandate removing sensitive data (tokens, production logs) and advise using temporary files or safe parameters to prevent shell interpolation when calling CLI tools.
- [COMMAND_EXECUTION]: The skill uses development tools like
git,gh,pnpm, andnxto automate repository contributions. The instructions are security-conscious, requiring explicit user authorization for write operations and providing guidance to avoid shell injection vulnerabilities. - [EXTERNAL_DOWNLOADS]: The skill interacts with the
nest-boot/nest-bootrepository on GitHub and utilizes standard package managers. These operations are restricted to the vendor's own ecosystem and represent normal development workflows for a maintenance agent.
Audit Metadata