mcpchecker MCP Evaluation - Report #414
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: mcpchecker MCP Evaluation - Report | |
| on: | |
| workflow_run: | |
| workflows: ["mcpchecker MCP Evaluation"] | |
| types: [completed] | |
| permissions: | |
| pull-requests: write | |
| actions: read | |
| jobs: | |
| report: | |
| name: Post Evaluation Results | |
| runs-on: ubuntu-latest | |
| if: github.event.workflow_run.event == 'issue_comment' | |
| steps: | |
| - name: Check if evaluation ran | |
| id: check | |
| env: | |
| GH_TOKEN: ${{ secrets.GITHUB_TOKEN }} | |
| run: | | |
| JOB_CONCLUSION=$(gh api "repos/${{ github.repository }}/actions/runs/${{ github.event.workflow_run.id }}/jobs" \ | |
| --jq '.jobs[] | select(.name == "Run MCP Evaluation") | .conclusion') | |
| if [[ -n "$JOB_CONCLUSION" && "$JOB_CONCLUSION" != "skipped" ]]; then | |
| echo "should-report=true" >> $GITHUB_OUTPUT | |
| else | |
| echo "should-report=false" >> $GITHUB_OUTPUT | |
| echo "Evaluation job was skipped, no report needed" | |
| fi | |
| - name: Download evaluation context | |
| if: steps.check.outputs.should-report == 'true' | |
| uses: actions/download-artifact@v8 | |
| with: | |
| name: eval-context | |
| path: eval-context/ | |
| run-id: ${{ github.event.workflow_run.id }} | |
| github-token: ${{ secrets.GITHUB_TOKEN }} | |
| - name: Post results comment | |
| if: steps.check.outputs.should-report == 'true' | |
| env: | |
| GH_TOKEN: ${{ secrets.GITHUB_TOKEN }} | |
| run: | | |
| PR_NUMBER=$(jq -r '.pr_number' eval-context/context.json) | |
| PR_SHA=$(jq -r '.pr_sha' eval-context/context.json) | |
| TASKS_PASSED=$(jq -r '.tasks_passed' eval-context/context.json) | |
| TASKS_TOTAL=$(jq -r '.tasks_total' eval-context/context.json) | |
| TASK_PASS_RATE=$(jq -r '.task_pass_rate' eval-context/context.json) | |
| ASSERTIONS_PASSED=$(jq -r '.assertions_passed' eval-context/context.json) | |
| ASSERTIONS_TOTAL=$(jq -r '.assertions_total' eval-context/context.json) | |
| PASSED=$(jq -r '.passed' eval-context/context.json) | |
| PASS_RATE=$(awk "BEGIN {printf \"%.1f\", $TASK_PASS_RATE * 100}") | |
| if [[ "${{ github.event.workflow_run.conclusion }}" != "success" ]]; then | |
| OVERALL="⚠️ Workflow failed" | |
| elif [[ "$PASSED" == "true" ]]; then | |
| OVERALL="✅ Passed" | |
| else | |
| OVERALL="❌ Failed" | |
| fi | |
| gh pr comment "$PR_NUMBER" --repo "${{ github.repository }}" --body "$(cat <<EOF | |
| ## mcpchecker MCP Evaluation Results | |
| **Commit:** \`${PR_SHA:0:7}\` | |
| **Summary:** ${TASKS_PASSED}/${TASKS_TOTAL} tasks passed (${PASS_RATE}%) | |
| | Metric | Result | | |
| |--------|--------| | |
| | Tasks Passed | ${TASKS_PASSED}/${TASKS_TOTAL} | | |
| | Assertions Passed | ${ASSERTIONS_PASSED}/${ASSERTIONS_TOTAL} | | |
| | Overall | ${OVERALL} | | |
| [View full results](https://github.com/${{ github.repository }}/actions/runs/${{ github.event.workflow_run.id }}) | |
| EOF | |
| )" |