Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Install and configure Grammarly SDK/CLI authentication. Use when setting up a new Grammarly integration, configuring API keys, or initializing Grammarly in your project. Trigger with phrases like "install grammarly", "setup grammarly", "grammarly auth", "configure grammarly API key".
.claude/skills/jeremylongshore-grammarly-install-auth/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -18% | 0% |
Configure Grammarly API authentication using OAuth 2.0 Bearer tokens. Grammarly provides three main APIs: Writing Score API, AI Detection API, and Plagiarism Detection API. All use the same auth pattern via api.grammarly.com.
scores-api:read, scores-api:writebash# .env (NEVER commit) GRAMMARLY_CLIENT_ID=your_client_id GRAMMARLY_CLIENT_SECRET=your_client_secret GRAMMARLY_ACCESS_TOKEN= # Populated after OAuth
typescript// auth.ts import 'dotenv/config'; async function getAccessToken(): Promise<string> { const response = await fetch('https://api.grammarly.com/ecosystem/api/v1/oauth/token', { method: 'POST', headers: { 'Content-Type': 'application/x-www-form-urlencoded' }, body: new URLSearchParams({ grant_type: 'client_credentials', client_id: process.env.GRAMMARLY_CLIENT_ID!, client_secret: process.env.GRAMMARLY_CLIENT_SECRET!, }), }); const { access_token, expires_in } = await response.json(); console.log(`Token obtained, expires in ${expires_in}s`); return access_token; }
typescriptasync function verify(token: string) { const response = await fetch('https://api.grammarly.com/ecosystem/api/v2/scores', { method: 'POST', headers: { 'Authorization': `Bearer ${token}`, 'Content-Type': 'application/json' }, body: JSON.stringify({ text: 'This is a test sentence for verification.' }), }); if (response.ok) console.log('Grammarly API connection verified'); else console.error('Verification failed:', response.status); }
| Error | Cause | Solution | |-------|-------|----------| | 401 Unauthorized | Invalid or expired token | Re-authenticate | | 403 Forbidden | Missing API scopes | Check enterprise admin settings | | invalid_client | Wrong credentials | Verify client ID and secret |
After auth, proceed to grammarly-hello-world.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-16 | fail→pass | 16,332 | 3,489 | -79% | 1 | 1 | 0% | 1,823 | 1,230 | -33% | 0 | 0 | — |
case-17 | pass→pass | 10,000 | 2,936 | -71% | 1 | 1 | 0% | 865 | 1,016 | +17% | 0 | 0 | — |
case-18 | pass→pass | 10,586 | 7,202 | -32% | 1 | 1 | 0% | 917 | 1,072 | +17% | 0 | 0 | — |
case-19 | pass→pass | 8,455 | 4,856 | -43% | 1 | 1 | 0% | 1,442 | 1,426 | -1% | 0 | 0 | — |
case-20 | pass→pass | 17,577 | 9,362 | -47% | 1 | 1 | 0% | 2,234 | 1,514 | -32% | 0 | 0 | — |
case-21 | pass→pass | 19,839 | 13,435 | -32% | 1 | 1 | 0% | 2,667 | 3,127 | +17% | 0 | 0 | — |
case-01 | fail→pass | 13,850 | 11,418 | -18% | 1 | 1 | 0% | 3,191 | 2,268 | -29% | 0 | 0 | — |
case-02 | fail→pass | 39,366 | 6,994 | -82% | 1 | 1 | 0% | 3,667 | 2,429 | -34% | 0 | 0 | — |
case-03 | fail→pass | 11,304 | 13,416 | +19% | 1 | 1 | 0% | 2,886 | 2,732 | -5% | 0 | 0 | — |
case-04 | fail→pass | 12,223 | 23,492 | +92% | 1 | 1 | 0% | 1,712 | 1,411 | -18% | 0 | 0 | — |
case-05 | fail→pass | 17,394 | 5,174 | -70% | 1 | 1 | 0% | 2,406 | 1,907 | -21% | 0 | 0 | — |
case-06 | pass→pass | 12,516 | 8,558 | -32% | 1 | 1 | 0% | 1,704 | 1,260 | -26% | 0 | 0 | — |
case-07 | pass→pass | 7,804 | 5,059 | -35% | 1 | 1 | 0% | 1,614 | 1,061 | -34% | 0 | 0 | — |
case-08 | fail→pass | 13,631 | 4,505 | -67% | 1 | 1 | 0% | 2,015 | 1,399 | -31% | 0 | 0 | — |
case-09 | pass→pass | 15,345 | 9,776 | -36% | 1 | 1 | 0% | 1,627 | 1,435 | -12% | 0 | 0 | — |
case-10 | pass→pass | 12,571 | 12,418 | -1% | 1 | 1 | 0% | 2,043 | 1,951 | -5% | 0 | 0 | — |
case-11 | pass→pass | 16,985 | 11,615 | -32% | 1 | 1 | 0% | 2,186 | 1,750 | -20% | 0 | 0 | — |
case-12 | fail→pass | 18,573 | 7,696 | -59% | 1 | 1 | 0% | 2,106 | 1,145 | -46% | 0 | 0 | — |
case-13 | pass→pass | 13,007 | 3,294 | -75% | 1 | 1 | 0% | 1,424 | 1,210 | -15% | 0 | 0 | — |
case-14 | fail→pass | 8,984 | 4,928 | -45% | 1 | 1 | 0% | 1,381 | 1,357 | -2% | 0 | 0 | — |
case-15 | fail→pass | 11,048 | 6,868 | -38% | 1 | 1 | 0% | 1,081 | 940 | -13% | 0 | 0 | — |
case-22 | pass→pass | 18,153 | 11,973 | -34% | 1 | 1 | 0% | 2,363 | 2,935 | +24% | 0 | 0 | — |
case-23 | pass→pass | 18,744 | 15,677 | -16% | 1 | 1 | 0% | 2,091 | 2,748 | +31% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +43 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.