forecast-skill

Provide pipeline deals with values, stages, and close dates from your CRM via MCP or a CSV/XLSX upload. The skill categorizes each deal, applies stage-based or confidence-based weighting derived from your historical win rates, and outputs a weighted forecast summary.

Use five evidence-based categories: Closed Won, Commit, Best Case, Pipeline, and Omit. Each deal is assigned via a decision tree requiring concrete evidence such as signed contracts, verbal commitments, confirmed budget, and decision-maker engagement.

Yes, the skill works without an integrated data source. Provide pipeline data directly in chat or upload CSV/XLSX files containing deal names, values, stages, close dates, and quota targets, and the same forecasting workflow applies.

Coverage ratio divides total active pipeline value by the quota target, with weighted and quality-adjusted variants that remove high-risk and omitted deals. The result is compared against your historical coverage-to-close ratio to judge sufficiency.

No, it never prescribes conversion rates or stage probabilities. All probabilities must be derived from your own historical closed-won and closed-lost data, or explicitly marked as assumptions requiring calibration against a historical baseline.

The forecast is a decision aid, not a final submission. Category assignments and risk assessments require validation by the rep and manager, and the skill cannot invent pipeline data or guarantee accuracy without quality historical baselines.