Agent Skill Library, kit
Granola to Claude: twelve skills that turn noisy calls into scored, cited records
Transcripts capture everything the buyer said and nothing about what you noticed. This stack adds the human layer back in, strips the noise, and turns each call into scored, cited records for deals, coaching, customer health, voice of customer, and follow-up. The download is the complete package: twelve skills, Granola Recipes, worked examples, and the class runbook.
Who it is for
Sales, customer success, product marketing, and RevOps leaders whose teams record calls in Granola and want evidence they can act on, not just summaries.
The download holds all 12 SKILL.md files plus templates and setup notes (25 files).
Safari note: Safari on Mac can unzip downloads automatically. To keep the kit zipped, open Safari, go to Settings, General, and turn off "Open 'safe' files after downloading". Chrome, Edge and Firefox keep the zip as it is.
Install this package
- Run the command below in your terminal. It downloads and extracts the package, then copies its skill folders into Claude Code's skills directory. Existing skill folders with the same names may be replaced.
- Keep the extracted
granola-gtm-call-stackfolder. Read its setup notes for any additional scripts, templates, or workspace steps before running the skills. For another assistant, follow its own skill installation instructions using the downloaded folders. - Start with the package's first step and check the output before using it.
curl -fL "https://www.therevenueaireport.com/agent-skills/kits/granola-gtm-call-stack/download.zip" -o "granola-gtm-call-stack.zip" && unzip -o "granola-gtm-call-stack.zip" && find "granola-gtm-call-stack" -name SKILL.md -print | while IFS= read -r file; do dir="$(dirname "$file")"; target="$HOME/.claude/skills/$(basename "$dir")"; mkdir -p "$target" && cp -R "$dir/." "$target/"; doneHub and setup
Route any call request to the right skill and build the context file every skill reads.
Capture and clean
Add what the rep noticed during the call, then strip the transcript down to an Evidence Ledger.
Granola Capture Markers
The human layer of the call stack. Teaches the shorthand reps type in Granola during a call (P pain, M moment, F fit, C confidence, D doubt, S stakeholder, X competitor, N next step) and parses those markers out of Granola notes into a Rep Read that AI can score.
Transcript Denoise
Strip a noisy sales, CS, or discovery call transcript down to an Evidence Ledger of atomic, quoted, source-tagged facts (pain, impact, moment, fit, stakeholders, process, competitors, objections, commitments, risks, product feedback) and a noise report.
Role scans
Read the ledger for deal qualification, coaching, customer health, voice of customer, and follow-up drafts.
Deal Qualification Scan
Score a deal from a call's Evidence Ledger against MEDDPICC (or SPICED) with the 0-2 rubric, a stakeholder grid, Green/Yellow/Red readiness, and the exact questions to ask next.
Call Coaching Scan
Coach the seller on a single call by stage (discovery, demo, pricing or proposal, negotiation, CS check-in) with a 0-4 rubric, talk-to-listen ratio, question quality, pains captured vs addressed, coded moments, and exact alternate wording.
Customer Health Scan
Read a customer call's Evidence Ledger for renewal risk, expansion signals, value realized, adoption, and sponsor health, and produce a Green/Yellow/Red health read, a 4-part executive brief (Context, Risks, Opportunities, Next Moves), and a customer recap with owners.
Voice of Customer Scan
Pull buyer language, pain themes, objections, competitor mentions, product feedback, and how reps are pitching from one call or a folder of calls, with cited quotes, plus content seeds (LinkedIn hook, email, ad line) written in the buyer's own words.
Follow up Builder
Turn a call's Evidence Ledger into a buyer follow-up email in the buyer's words, an internal Slack recap, and proposed CRM field updates, all as drafts for human review.
Across many calls
Run one Recipe across a Granola Folder, update the ICP from win and loss patterns, and build a scored Account Context.
Folder Recipe run
Run one Recipe (a fixed extraction schema) across every meeting in a Granola Folder and return one table with one row per call, with cited quotes and confidence tags. Works two ways, inside Granola (paste the Recipe into Granola chat on the Folder) or in Claude Code over the Granola MCP.
Win Loss ICP Update
Run the win/loss Recipe across Granola Folders of won and lost deals, compute which patterns show up more in wins than losses (lift, with counts), check how well reps' live confidence predicted outcomes, and propose evidence-backed changes to the ICP and MOMENT signal list in gtm-context.md.
Account Context Builder
Pull a Granola meeting into Claude over MCP (typed notes plus transcript) and build or update a scored Account Context with FIT, PAIN, MOMENT, CONFIDENCE, a bucket, a hypothesis, why we might be wrong, and one next step. The rep's typed confidence from the call is carried in as-is.
How to start
- 1. Download the zip and unzip it.
- 2. Claude Code: copy the skill folders with cp -r skills/* ~/.claude/skills/ (or into .claude/skills in one project).
- 3. Connect Granola: claude mcp add granola --transport http https://mcp.granola.ai/mcp, then type /mcp and authenticate.
- 4. Say: set up my GTM context. Answer three questions. Then say: run the call stack on my last call.
- 5. Claude desktop or web: connect Granola from connectors, zip each folder under skills/, and upload it in skills settings.
- The download is the original zip, including granola-recipes, examples, and CLASS-RUNBOOK.md.
- On Granola's free plan the MCP returns notes from the last 30 days only. Raw transcripts and Folder listing need a paid plan.
- Nothing is sent and no CRM is written. A human decides.
Built by Jonathan Kvarfordt on the GTM AI OS. Supplied as a complete zip and served unchanged.