tune-revops
Skill1 starsChanged 3 months agoMITView on GitHub
Build or audit the revenue plumbing — CRM configuration, stage definitions, lead scoring, response SLAs, data hygiene, and forecast mechanics. Run when the pipeline is first outgrowing a markdown file, quarterly thereafter, or whenever the same operational problem keeps recurring.
In this repository22 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- build-funnelSkill
- close-dealSkill
- connect-toolsSkill
- create-contentSkill
- define-positioningSkill
- design-offerSkill
- experimentSkill
- grow-referralsSkill
- grow-seoSkill
- health-checkSkill
- launch-campaignSkill
- launch-productSkill
- manage-pipelineSkill
- measureSkill
- plan-strategySkill
- research-marketSkill
- review-monthlySkill
- run-adsSkill
- run-outboundSkill
- run-partnershipsSkill
- setup-teamSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Define the stages: Every stage in knowledge/pipeline.md gets an objective exit criterion two people would agree on.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/tune-revops — Fix the Plumbing
Inputs: knowledge/pipeline.md, knowledge/business-context.md (volume and model), knowledge/results-history.md (closed-deal history), knowledge/tool-registry.md (is a CRM connected?), deal files in knowledge/deals/.
Agents: revops-operator leads; tools-operator executes CRM changes; growth-analyst validates the forecast math; sales-developer and deal-closer confirm the stage definitions match how selling actually happens.
Method: knowledge/methods/revenue-operations.md.
Steps
- Right-size check. Current monthly lead volume vs. the machinery in place (method §1). Recommend upgrade, hold, or — often — simplify. A CRM the operator doesn't open is worse than a markdown table they do.
- Define the stages. Every stage in
knowledge/pipeline.mdgets an objective exit criterion two people would agree on. Sample five real deals per stage and check them against the definition; misplacements reveal which definitions are vague. - Set the field discipline. Confirm the six required fields (source, stage, amount, expected close, next action + date, owner) exist and are populated. Kill any field nobody uses; name the decision each surviving field informs.
- Build or validate lead scoring. Fit and intent scored separately (method §4). First model comes from the last 20 closed deals — what separated wins from losses. If a model already exists, test it: do high-scored leads actually close better? If not, rebuild rather than defend.
- Set response SLAs and routing. Per lead type, with the notification path that makes the SLA achievable (Slack, email, wherever the operator actually looks). Measure the current median response time first — the gap between assumed and actual is usually the finding.
- Run the hygiene sweep. Deduplicate (merge, never delete), fix bounces, fill or flag blank required fields, age out 12-month-dormant contacts into a dormant segment, verify suppression lists propagate across every system and channel.
- Recompute the numbers. Stage-to-stage conversion rates from real history, written back to
knowledge/pipeline.md. Score last period's forecast accuracy against actuals and state the error plainly. - Hunt the leak. The highest-value output of this skill: which stage has the worst conversion and the longest dwell time? That's where leads die. Name it, quantify it, and recommend the one fix that recovers the most — usually a definition or an SLA, rarely a tool.
Before you start
- Required inputs:
knowledge/pipeline.mdatseeded+ with some real history. With zero closed deals, skip scoring and forecasting entirely — do steps 1–3 and 6 only, and say why. - Scan
knowledge/decision-log.mdfor tagpipeline— stage definitions and disqualification rules are recorded decisions; changing one is a new decision, not a tweak. - Gate check: none for internal configuration. Any automation that sends to contacts (auto-replies, nurture triggers) falls under the outbound gate before it goes live.
If things go wrong
- CRM unauthorized or unavailable → do the definition work anyway (it's most of the value) and park the configuration steps in
knowledge/pending.mdwith the authorization instructions. - Too little history to validate scoring or compute conversion rates → say so and use labeled conservative defaults; a stated guess is honest, a computed-looking guess is not.
- Stage definitions the sales agents disagree with → resolve it now, in this run. A definition the people using it reject will be ignored within a week.
Outputs
Documented stage definitions with exit criteria, a validated (or newly built) scoring model, response SLAs, a cleaned database, recomputed conversion rates, and the named leak with its fix. Close with the report per templates/skill-report.md.
Writeback
knowledge/pipeline.md (stage definitions, conversion rates, hygiene results); CRM configuration state to knowledge/tool-registry.md; stage/scoring/SLA decisions (tag: pipeline) to knowledge/decision-log.md; forecast-accuracy score to knowledge/results-history.md; a learning line to knowledge/methods/revenue-operations.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

