jdconley/run-live-evals
.cursor/skills/run-live-evals/SKILL.md
Run Wodbrains worker parse evals and live Gemini tests locally using Wrangler `.dev.vars` keys. Use when you need to run `parse.evals.test.ts` / `parse.gemini.test.ts` against the real model (RUN_LIVE_AI_TESTS=1).
$ install --global
skillsauthnpx skillsauth add jdconley/wodbrains run-live-evalsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 16, 2026, 11:33 PM344.9s1 file scanned
SKILL.md
- name:
- run-live-evals
- description:
- Run Wodbrains worker parse evals and live Gemini tests locally using Wrangler `.dev.vars` keys. Use when you need to run `parse.evals.test.ts` / `parse.gemini.test.ts` against the real model (RUN_LIVE_AI_TESTS=1).
Run live evals (worker)
What this runs
apps/worker/test/parse.evals.test.ts(the eval manifest / fixtures)apps/worker/test/parse.gemini.test.ts(text + URL + image live parsing)
These are “live” because they call the real Gemini API using GOOGLE_GENERATIVE_AI_API_KEY.
Prereqs
- Ensure
apps/worker/.dev.varsexists and containsGOOGLE_GENERATIVE_AI_API_KEY=... - Do not commit
.dev.vars(it contains secrets)
Run
From repo root:
pnpm test:live-evals
Equivalent manual command:
set -a && . "apps/worker/.dev.vars" && set +a && RUN_LIVE_AI_TESTS=1 \
pnpm --filter worker test -- test/parse.evals.test.ts test/parse.gemini.test.ts
Common gotchas
- Tests “skipped”: live suites are gated on
RUN_LIVE_AI_TESTS=1and a present API key. - Workers runtime env: in
@cloudflare/vitest-pool-workerstests, prefercloudflare:testenvbindings for gates instead of relying onprocess.env.
Related Skills
jdconley/ui-guidelines
documentation
VerifiedTrustedCommunity
UI design guidelines for WOD Brains app - mobile-first, app-like design with consistent patterns. Use when making UI changes.
1SKILL.mdUpdated Apr 16, 2026
jdconley/test-on-change
testing
VerifiedTrustedCommunity
Always create and run tests affected by changes, including Playwright for UI changes. Use when modifying Wodbrains features or UI.
1SKILL.mdUpdated Apr 16, 2026
jdconley/readme-workflow
documentation
VerifiedTrustedCommunity
Regenerate README screenshots and the demo video for WOD Brains.
1SKILL.mdUpdated Apr 16, 2026
jdconley/og-image-workflow
testing
VerifiedTrustedCommunity
Generate and update the WOD Brains OG image PNG/JPG using the mascot SVG and a Playwright-rendered layout. Use when changing OG image copy or layout.
1SKILL.mdUpdated Apr 16, 2026