Back to directory
Frontendstackforgeai/copilot-evals
@stackforgeai/copilot-evals
LLM evaluation and quality assurance framework for GitHub Copilot SDK — judge-based scoring, prompt injection detection, benchmark suites, and CI/CD eval pipelines, all routed through copilot-guard.
Suggested install command
npx skills add stackforgeai/copilot-evals/@stackforgeai/copilot-evalsAlways inspect the linked repository and skill instructions before running commands. Skills are instructions; permissions and execution still matter.
Compatibility
Agent support matrix
3 supported
| Agent | Status |
|---|---|
| Claude Code | Supported |
| OpenCode | Not listed |
| Cursor | Not listed |
| MCP | Not listed |
| GitHub Copilot | Supported |
| Windsurf |