GTM Vault skills
Skills built from systems that already ran.
Every operator skill here is built from a system a named operator walked through on GTM Vault, with the episode, the timestamps and the number it produced. Download one into Claude, or run it through the GTM Vault MCP. Free for everyone. 14 operator skills and the 8 skills GTM Vault itself runs on.
Built from the systems of
22 of 22 skills
SignalsFree
PQL scoring and routing agentScores every product signup out of 100 from product and CRM data, then routes it to sales or nurture.
7 signals, scored out of 100Michael Bartimer scores each PLG signup on seven signals out of 100, about half pulled live from PostHog, with 60 or greater sales ready.
SignalsFree
Review intelligence signal layerTurns public reviews into buying signals: a star filter before the model, then Claude flags fixable complaints.
16 of 20 reviews dropped before the modelIn the test run Vinayak Mishra showed, the star-rating gate discarded 16 of 20 reviews and passed 4 to Claude.
ProspectingFree
Account monitoring universeScores a TAM on four layers, watches it for signals, and gates outreach from first signal to phone call.
17 to 18 days down to the same dayUmar Farooq Adam said inbound lead time fell from 17 to 18 days to the same day once the universe was live.
ResearchFree
Win-loss gap signal discoveryMines won against lost accounts for fit and anti-fit signals, scores targets, and stages gated outbound.
Under 30 minutes, under $10Jai Toor said a month-long workflow that cost almost $2,000 now runs in under 30 minutes for under $10.
OutreachFree
Outbound run from Claude CodeScrape, enrich, personalize and launch split-tested outbound campaigns from Claude Code prompts.
8 opportunities from 180 emailsTim Scheuer reported eight positive sales opportunities from just over 180 emails, set up in about 40 minutes of prompting.
OutreachFree
Cold email as a GTM engineA lean cold email system: a cheap sending stack, layered ICP filters, a two-step offer and an AI post-call closer.
About $200 a monthCharles Cormier said he would need about 200 dollars a month to rebuild his whole cold email stack.
EnrichmentFree
MCP and API enrichment pipelineSplits enrichment between MCP context checks and direct API volume calls, run by a supervisor agent.
About 50 cents a leadAlexander Shartsis said Skyp runs managed enrichment at around 50 cents a lead after choosing which APIs it calls.
SalesFree
Agents behind the repHides an orchestrator and single-job outbound agents behind one rep action, such as a CRM checkbox.
About 60% of the BDR workflowAngelo Quiambao said his agents became around 60% of the daily BDR workflow at Qualified.
DealsFree
Revenue intelligence layerScores target accounts and inspects pipeline against segment win rates, cycle lengths and stage gates.
5,000 scored MQAsMichele Shepard said Demandbase produced 5,000 scored MQAs at Emburse and saved 20 minutes per account.
AEOFree
Get cited by AI searchPlan and measure how a B2B company gets seen, cited and chosen in ChatGPT, Perplexity and Claude.
3 weeks to a monthJason Patel said to give it three weeks to a month before visibility shows, far quicker than SEO.
SignalsFree
Account scoring to a BDR action layerScore the TAM in Clay, monitor signals only on accounts that clear the score, and give BDRs one action layer on the CRM.
30 to 60 minutes a dayGarrett Wolfe said the action layer saves reps as little as half an hour to an hour a day across his clients.
EnrichmentFree
Enrichment actions in the rep's browserBring RevOps-built Clay workflows into the rep's browser as one-click actions that read the page, query Clay or the CRM, and write back.
SalesFree
Headless dashboard for AEsOne custom AE view joins CRM, data, outreach and back office sources, with engagement and call context written back automatically.
OutreachFree
LinkedIn outbound with Loom demosFounder-led LinkedIn outbound to a known ICP, with one-minute personalized Loom demos and technical Reddit posts that build demand.
Content ops · PublishFree
Episode write-up formatThe canonical structure for every episode post, YouTube description, chapters, SEO fields and email header.
How GTM Vault is run · Needs: A transcript or recording notes. Substack and YouTube.
Content ops · PublishFree
LinkedIn episode announcementThe default launch post for every new episode, written to carry the trailer.
How GTM Vault is run · Needs: The episode write-up and the trailer.
Content ops · PublishFree
Clip caption packOne clip transcript in, captions for Instagram, LinkedIn, YouTube Shorts, X and Substack Notes out.
How GTM Vault is run · Needs: A clip transcript.
Content ops · ProduceFree
Trailer brief for the editorA per-episode creative brief so the editor hits the trailer on draft one, plus the review of the cut.
How GTM Vault is run · Needs: The episode audio, video or transcript.
Content ops · ProduceFree
Episode thumbnail railThe YouTube thumbnail and Substack card for every episode, built from Canva masters.
How GTM Vault is run · Needs: Canva and one master design per show and surface.
Content ops · ProduceFree
Post cover imagesBranded cover and OG images for essays and laws, rendered from HTML templates.
How GTM Vault is run · Needs: Python with Playwright. Templates are included.
Content ops · GrowFree
Signals digest railEvery two weeks: new GTM and AI studies, raises, open roles and events, as a digest and an essay footer.
How GTM Vault is run · Needs: Web search.
Content ops · GrowFree
AI citation railKeeps the Stack Index citable by AI answer engines, and measures it with a fixed prompt panel.
How GTM Vault is run · Needs: A public index or library you want cited.
How these are built
- One system, one operator, one episode. Each operator skill starts from a system described on the record and cites it by timestamp. The operator did not write or review the skill; it is built from what they said, and their numbers are theirs, not a promise.
- Runnable, not a summary. Every skill ends in steps Claude follows on your company and an artifact it produces: a scoring sheet, a spec, a sequence, a measurement loop.
- Tools are evidence, not placement. Each skill lists the tools the operator runs, from the Stack Index. Any tool doing the same job works. Placement is never for sale.
Two ways to use them
- Install one. Download the zip and add it as a skill in Claude's settings. In Claude Code, unzip it into
~/.claude/skills/. - Run it from the GTM Vault MCP. All 22 sit in the Skills group next to the thirty-four playbooks, free for every GTM Vault subscriber. Connect GTM Vault to Claude, then ask:Run the GTM Vault PQL scoring agent skill on our signups.