Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Uses MCP Connectors to read Gmail inbound leads, score them by ICP fit, draft personalized responses, and log qualified leads to your CRM. Turns your inbox into an automated pipeline.
.claude/skills/onewave-ai-gmail-to-crm-pipeline/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 95% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -29% | 0% |
Turn a Gmail inbox into a structured sales pipeline using Claude's MCP Connectors -- the Gmail connector reads and drafts email, and the Supabase connector persists CRM data. No external scripts or API keys required.
pipeline_config table. See references/configuration.md (first run, error handling) and references/crm-schema.md.references/gmail-retrieval.md.references/gmail-retrieval.md.null rather than guessing. See references/lead-extraction.md.references/scoring-model.md.references/response-templates.md.references/crm-schema.md.lead-pipeline-report.md in the working directory, then display the executive summary. See references/pipeline-report.md.Between runs, handle manual lead commands (mark contacted, move stage, disqualify, add note, schedule follow-up, query leads) and ICP/search customization. See references/configuration.md.
references/gmail-retrieval.md -- Gmail MCP tools, search queries, parsing rules, extracted field set.references/lead-extraction.md -- Lead profile JSON schema and extraction guidelines.references/scoring-model.md -- ICP, intent, urgency scoring tables; tier mapping; adjustment rules.references/response-templates.md -- Per-tier response templates, personalization rules, anti-patterns, draft creation.references/crm-schema.md -- Supabase MCP tools, table DDL, logging procedure and SQL.references/pipeline-report.md -- Report Markdown structure and reporting SQL queries.references/configuration.md -- First-run setup, customization, manual commands, error handling, privacy, invocation triggers.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | fail→pass | 10,810 | 8,789 | -19% | 1 | 1 | 0% | 1,647 | 1,991 | +21% | 0 | 0 | — |
case-01 | fail→fail | 16,778 | 16,422 | -2% | 1 | 1 | 0% | 3,282 | 1,379 | -58% | 0 | 0 | — |
case-02 | fail→fail | 20,700 | 3,003 | -85% | 1 | 1 | 0% | 3,546 | 1,171 | -67% | 0 | 0 | — |
case-03 | fail→fail | 24,868 | 3,051 | -88% | 1 | 1 | 0% | 4,666 | 1,067 | -77% | 0 | 0 | — |
case-04 | pass→pass | 15,373 | 4,166 | -73% | 1 | 1 | 0% | 2,757 | 1,439 | -48% | 0 | 0 | — |
case-05 | fail→fail | 9,973 | 16,468 | +65% | 1 | 1 | 0% | 1,729 | 3,436 | +99% | 0 | 0 | — |
case-06 | fail→pass | 8,966 | 4,415 | -51% | 1 | 1 | 0% | 1,392 | 1,350 | -3% | 0 | 0 | — |
case-08 | fail→pass | 4,565 | 4,514 | -1% | 1 | 1 | 0% | 730 | 1,424 | +95% | 0 | 0 | — |
case-09 | fail→pass | 13,240 | 2,339 | -82% | 1 | 1 | 0% | 2,239 | 1,043 | -53% | 0 | 0 | — |
case-10 | fail→pass | 8,553 | 1,680 | -80% | 1 | 1 | 0% | 1,338 | 950 | -29% | 0 | 0 | — |
case-11 | fail→fail | 9,293 | 2,498 | -73% | 1 | 1 | 0% | 1,279 | 1,019 | -20% | 0 | 0 | — |
case-12 | pass→pass | 15,104 | 12,072 | -20% | 1 | 1 | 0% | 2,582 | 2,426 | -6% | 0 | 0 | — |
case-13 | fail→fail | 13,056 | 12,433 | -5% | 1 | 1 | 0% | 2,531 | 2,792 | +10% | 0 | 0 | — |
case-14 | fail→pass | 13,760 | 5,398 | -61% | 1 | 1 | 0% | 2,044 | 1,511 | -26% | 0 | 0 | — |
case-15 | pass→pass | 5,995 | 4,162 | -31% | 1 | 1 | 0% | 1,000 | 1,447 | +45% | 0 | 0 | — |
case-16 | fail→pass | 13,399 | 6,285 | -53% | 1 | 1 | 0% | 2,569 | 1,601 | -38% | 0 | 0 | — |
case-17 | pass→pass | 13,082 | 7,692 | -41% | 1 | 1 | 0% | 2,105 | 2,013 | -4% | 0 | 0 | — |
case-18 | fail→fail | 7,243 | 3,489 | -52% | 1 | 1 | 0% | 1,100 | 1,224 | +11% | 0 | 0 | — |
case-19 | pass→pass | 11,671 | 5,147 | -56% | 1 | 1 | 0% | 1,802 | 1,434 | -20% | 0 | 0 | — |
case-20 | pass→pass | 9,264 | 5,176 | -44% | 1 | 1 | 0% | 1,576 | 1,475 | -6% | 0 | 0 | — |
case-21 | pass→pass | 11,442 | 12,503 | +9% | 1 | 1 | 0% | 2,636 | 3,563 | +35% | 0 | 0 | — |
case-22 | pass→pass | 5,150 | 7,319 | +42% | 1 | 1 | 0% | 990 | 2,043 | +106% | 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. 22 cases were attempted. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.