Open-source GTM intelligence and automation for any sales call transcript.
Turn sales conversations into evidence-bound coaching, deal scoring, and automated CRM updates.
GTM Superintelligence is an open-source framework that processes sales call transcripts into structured coaching reports, health scores, and post-call workflows. It is built for sales leaders, operations teams, and reps who want transparent, editable scoring rubrics instead of black-box vendor tools.
It ingests transcripts from any recorder, classifies call types, infers desired outcomes, scores interactions against YAML rubrics, and runs 30 post-call agents. The pipeline handles everything from single-call coaching to multi-call deal and account health scoring, plus CRM auto-fills.
It suits GTM engineers and sales leaders who want to customize coaching rubrics and run local Python CLIs or Claude Code skills. It does not suit teams looking for a zero-setup, turnkey SaaS product with native web recording out of the box.
Normalizes transcripts from Attention, Gong, Fireflies, Otter, Zoom, Recall.ai, Grain, VTT, SRT, or text files.
Call types, sales frameworks, and scorecards are plain YAML files you can read, diff, and fork.
Every score and piece of feedback ties directly to a verbatim quote from the transcript.
Includes production agent templates that draft follow-ups, flag single-threaded deals, and run post-mortems.
Run the entire analysis directly inside Claude Code without needing an API key.
Free under its Apache-2.0 license. The only operational cost is running it yourself via the Anthropic API or Claude Code.
873 open remote AI-training roles, updated November 15, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
Yes. The tool features adapters for Attention, Gong, Fireflies, Otter, Zoom, Recall.ai, Grain, VTT, SRT, and plain text files.
You need an Anthropic API key to run the Python CLI, but you can also run the entire tool natively inside Claude Code without an API key.
Scoring uses public methodologies like SPICED, MEDDPICC, and BANT stored as editable YAML knowledge base files.
See Zekai first in Google