Audit lifecycle definitions, scoring, routing, pipeline controls, data quality, handoff SLAs, and revenue metrics without smoothing over missing evidence.
A tidy funnel chart can hide incompatible denominators, silent routing fallbacks, duplicated events, mixed currencies, or a stage definition that changed without an effective date. Those are control problems, not dashboard formatting problems.
This skill turns CRM exports, automation rules, funnel reports, and team definitions into an evidence-bound revenue operations audit. It maps which system owns each field, keeps contacts and opportunities at the right grain, reconciles stage movement, tests scoring boundaries and routing conflicts, rebuilds metric definitions, and records every unresolved exception with an owner and closure requirement.
Use it before changing lead scoring, routing, lifecycle automation, pipeline reporting, or a board metric. It can also review a suspected handoff leak or a dashboard disagreement. The output separates observed evidence, stakeholder claims, inferences, and proposals, then returns one of four explicit verdicts: effective, effective with exceptions, not effective, or insufficient evidence.
Version 1.0.0 was evaluated with gpt-5.6-sol in 12 fresh read-only contexts. The candidate scored 100 across three cases, compared with 83.33 for the same agent without the skill and 78.75 for the exact free upstream. The cases covered funnel reconciliation, scoring and routing launch controls, and pipeline/SLA board metrics. This evaluation measures the supplied cases, not live CRM correctness or buyer outcomes.
The package includes a local checker for the optional canonical JSON audit record. Passing it confirms structure, references, reconciliation arithmetic, and verdict precedence. It does not prove that an export is complete or that a production control worked. Maintenance and install support are handled by the platform-operated publisher for this listing.
Input
Audit June funnel reporting. The CRM counts distinct contacts, marketing counts MQL events including recycled leads, and sales defines SQL as an opportunity with completed discovery. Stage movement does not reconcile by one contact.
Output
A blocked reporting verdict; source and grain map; opening-plus-entries-minus-exits reconciliation; duplicate and recycle exceptions; effective-date and ownership gaps; reproducible metric definitions; control tests; and exact evidence needed before capacity planning.
Input
Review an enterprise lead launch where named accounts overlap territories, inactive owners remain eligible, no fallback exists, 38 scores differ between systems, and the test export contains only final owner.
Provide the decision the audit must support, review period, systems in scope, current definitions and rules, and the best available CRM, marketing automation, routing, pipeline, SLA, or dashboard exports. Include extraction times, filters, keys, grains, timezone, currency, and known limitations when available. Ask for a complete audit or name one control family. The skill will not mutate production systems; it returns findings, evidence requests, proposed changes, verification cases, and closure gates for review.
The decision owner, review period, go-to-market motion, segments, systems, timezone, currency, exclusions, and the decision this audit must support.
Lifecycle and opportunity-stage definitions, scoring components, routing precedence, handoff SLAs, metric formulas, owners, and effective dates when known.
Dated exports, reports, rule snapshots, transition or assignment logs, filters, grains, stable keys, row counts, and known limitations. Redact personal contact data where possible.
Who can approve definitions or production changes, plus any test environment, rollout, monitoring, and rollback constraints. The skill prepares changes but does not execute them.
SKILL.md; revenue operations audit rubric; report and canonical JSON output contract; read-only JSON audit checker; Agent Skills interface metadata
No reviews yet.
Output
A not-effective launch verdict; classified score differences; threshold and override evidence gaps; routing conflict, no-match, availability, duplicate, retry, and fallback cases; safe rollout and rollback gates; prioritized owners and closure requirements.
Input
Audit a pipeline dashboard that adds USD, EUR, and GBP nominally, omits stage history, and reports 11-minute speed-to-lead after excluding 180 of 1,000 submissions with missing first-contact timestamps.
Output
An insufficient-evidence or not-effective verdict; currency and authority gaps; CRM-to-billing and billing-to-CRM tests; duplicate, orphan, credit-note, stage-skip, and close-date exceptions; denominator reconstruction; metric dictionary; and closure evidence for controlled board reporting.
Trebuh
Updated Jul 13, 2026
Automated scan passed. It found 1 finding; highest severity info.
Scanned Jul 16, 2026.
Automated scanning reduces risk but cannot guarantee a package is safe for every environment.
Creator
Mmovaro7
