Win-loss review

State: to-be-approved. Deploy-verified against a live workspace: not yet. Treat Done when below as the acceptance test and review cargo-ai cdk plan before deploying. Make no outcome claim for this skill until it is approved.

The outcome

What the CRM knows about who buys and who does not stops living in a report nobody reopens. Once a month one agent runs, and two things land in your repository:

1. The context files. On the first pass the agent verifies icp/ against what separates won from lost and names a disqualifier; it writes dated insight/ files in three buckets (who we talk to, where we win, where we lose), every count with its denominator; objection/ files from lost reasons where the CRM records them; and a client/ file per closed-won account. Every month after, it appends what the new deals say and edits nothing that exists.
2. The run record. outputs/<date>-win-loss-review/README.md holds every query's result, counts and ids only, so every [R: <query>] tag in the context files points somewhere a reader can check. Its date is where next month starts.