qualify-pipeline-check · git:20260810.5591440 · 2026-08-10 · sha256 d6b09005064bb513
qualify-pipeline-check git:20260810.5591440A
Immutable. This exact content never changes and is served at /api/v1/blob/d6b09005064bb513.
--- name: qualify-pipeline-check description: Sweeps a CRM/pipeline export for deals that advanced past the Qualify stage without the facts that stage is supposed to establish — economic buyer, pain, budget signal, a close date consistent with the procurement cycle, a next step, and a qualification-framework record. Use when RevOps or a manager wants to audit the Qualify pipeline for deals advanced on hope rather than evidence. Does not read call transcripts — that's `meeting-to-qualify` (audits the qualification call itself) or `sql-to-qualify` (audits the SQL handoff call); this one only asks whether the CRM record itself can prove qualification happened. license: MIT metadata: zime:category: new-business zime:dimension: stage zime:stage: Qualify pipeline zime:input-modes: csv,connector --- # Qualify Pipeline Check Sweeps a deal/pipeline export for rows in the Qualify stage that lack the facts Qualify is supposed to establish before a deal is allowed to advance. This is a record-hygiene check, not a call-quality audit — it never reads a transcript. Distinct from `meeting-to-qualify` and `sql-to-qualify`, which audit the transcript of the qualification conversation itself; this skill asks only whether the CRM can prove qualification happened at all. ## When to use this - RevOps wants a periodic sweep of the Qualify stage for deals that drifted forward without the underlying facts ever being captured. - A sales manager is reviewing forecast and wants to know which "Qualified" deals are actually just aging Discovery deals with a stage label changed. - Before a pipeline review, to flag deals that will not survive real scrutiny on economic buyer, budget, or next step. ## Input A `.csv` deal/pipeline export. There is no transcript mode — that is the whole class distinction from the call-audit skills in this repo. Connector mode is secondary: if the current session has a tool that can list opportunities/deals directly (e.g. a CRM MCP tool), call it and treat each returned row exactly like a CSV row. Otherwise, ask for a CSV export. ``` claude "run qualify-pipeline-check on ./exports/pipeline.csv" ``` ## Column detection Match headers case-insensitively, ignoring `_`/`-`/space differences. Accept these synonyms per concept (see `references/rubric.md` for the full list per flag): - **stage**: `stage`, `deal_stage`, `dealstage`, `pipeline_stage` - **close date**: `close_date`, `closedate`, `expected_close`, `expected_close_date` - **deal value**: `amount`, `deal_value`, `value`, `arr`, `deal_size` If a column a check needs is absent from the export entirely, that check reports **Unknown (column missing)** for every row — state this once, up front, in the output, and never infer the missing value from another column. ## Stage filter Operate only on rows whose stage column fuzzy-matches "Qualify" (e.g. `Qualify`, `Qualified`, `Qualification`). Report at the top of the output how many rows were in scope out of the export's total row count. ## Checks Run every check in `references/rubric.md` against every in-scope row. Each check returns one of three states per row: **Flagged** / **Clean** / **Unknown**. **Evidence rule.** Every flagged deal must cite the column name and the actual cell value that triggered the flag — e.g. `close_date = 2026-01-14` or `champion = (empty)`. A flag with no cited cell does not ship. ## Output One markdown table, flagged deals first, ordered by most flags first: | Deal | Flags | Evidence | Suggested action | Then exactly two closing lines: - `N of M deals in Qualify pipeline flagged` - The single most common flag across the flagged deals Nothing else — no scores, no letter grades, no percentages invented from data that isn't there. ## Sample data `assets/sample-pipeline-qualify-pipeline-check.csv` is a synthetic pipeline export with 10 Qualify-stage rows — several deliberately triggering flags and at least two genuinely healthy deals that should come back Clean on every check. Run against it first. ## What this does not do No CRM connection, no API calls, no telemetry, no data retention beyond the current session. It reads the file you point it at (or the connector rows you hand it) and nothing else.