Discovery Call to CRM

Every deal that stalls after a good discovery call is a follow-up problem, not a product problem.

Discovery Call to CRM is an OpenClaw skill that processes a raw transcript and produces two things: a structured CRM note aligned to MEDDIC, SPICED, or BANT, and a follow-up email that quotes three specific phrases the prospect actually said. The whole pipeline takes under two minutes. The gap between hanging up and hitting send shrinks from hours to seconds.

What it does

• Extracts qualification signals (pain, metrics, economic buyer, decision process, timeline, champion) from any transcript using phrase matching
• Flags missing qualification fields and generates specific questions to ask on the next call
• Renders a paste-ready CRM note in Salesforce, HubSpot, or plain markdown format
• Drafts a follow-up email with three cited prospect quotes and one concrete proposed next step

What's in the package

All files listed below are bundled as fenced code blocks inside SKILL.md — there's no separate folder to install. Copy each code block into its own file using the path shown above it.

• SKILL.md — the full workflow your agent follows.
• scripts/extract_signals.py — parses a transcript and produces structured signals JSON with qualification score.
• scripts/render_crm_note.py — formats signals as a CRM-paste-ready note (Salesforce, HubSpot, or markdown).
• scripts/draft_followup.py — drafts a follow-up email with cited quotes and a single next step.
• data/qualification-frameworks.yaml — full field definitions for MEDDIC, SPICED, and BANT.
• scripts/examples/ — sample transcript, signals JSON, CRM note, and follow-up email so you can verify the pipeline end to end.

How it ships

Drop SKILL.md into ~/.openclaw/skills/discovery-call-to-crm/. Restart OpenClaw. Run /discovery-call-to-crm or paste a transcript and say 'process this discovery call.'

Questions: hackedcubeapps@gmail.com (response within 48 hours).