quality-gates-enforcer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (LOW): The skill references third-party GitHub Actions (
romeovs/lcov-reporter-action,andresz1/size-limit-action) that are outside the trusted organization list. While common in CI/CD workflows, they represent unverifiable external dependencies. - [PROMPT_INJECTION] (LOW): Identifies an indirect prompt injection surface where tool outputs (lint results, coverage reports) are ingested and then published as PR comments via
actions/github-script. - Ingestion points:
coverage-summary.jsonand tool stdout (fromnpm run lintandnpx tsc). - Boundary markers: Absent; no delimiters are used to wrap ingested content.
- Capability inventory:
github.rest.issues.createComment(write access to PR metadata). - Sanitization: Tool output is read directly from local files or captured stdout and posted without filtering or escaping.
Audit Metadata