call-to-crm
Skill0 starsChanged 29 days agoMITView on GitHub
Turns a sales call transcript or rough notes into a structured CRM update, a MEDDICC summary with gaps, a handoff note for the account executive and a short follow-up email. Never invents facts that are not in the notes. Use when the user pastes a transcript or call notes and asks to "log this", "summarise the call", "write the recap", "update the CRM", "prepare the handoff", or "draft the follow-up".
In this repository4 other files it gives its agents.
- discovery-call-prepSkill
- objection-handlerSkill
- sales-routerSkill
- signal-verified-outreachSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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
- Distinctive: line 11
Never infer, only record when explicitly said: budget, authority, timeline, pain severity, competitor, buying intent, next-step owner.
- Distinctive: line 14
Every field is filled from the transcript or notes, or marked unknown.
- Very specific: line 27
If a prep brief from discovery-call-prep exists, compare: which hypotheses were confirmed, rejected or not tested?
≈ “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
Call → CRM
Turn a messy call into four clean outputs, using only what was actually said.
No-invention list
Never infer, only record when explicitly said: budget, authority, timeline, pain severity, competitor, buying intent, next-step owner. If it wasn't said, it is Unknown.
Rules
- Every field is filled from the transcript or notes, or marked unknown. Never infer a budget, a decision maker or a date that wasn't stated.
- Quote the prospect's own words for pain and impact when possible. Their words are stronger than a paraphrase.
- Tag every line as Said (they said it, with a quote where possible), Interpreted (your read, labelled) or Unknown (not discussed). An interpretation must never reach the CRM as a fact.
Workflow
1. Extract
- Attendees and roles.
- Current situation, pain points (with quotes), impact, desired outcome.
- Decision process, other stakeholders, timing, budget signals.
- Objections raised and how they were left.
- Agreed next steps, owners, dates.
If a prep brief from discovery-call-prep exists, compare: which hypotheses were confirmed, rejected or not tested?
2. Score qualification (MEDDICC)
For each letter: what we know, the evidence (quote or note), and what is still missing. End with the top 3 gaps to close in the next conversation.
3. Write the outputs
A. CRM update
Stage: <suggested stage + why>
Next step: <action> — owner — date
Pain (in their words): "<quote>"
Stakeholders: <name, role, stance if known>
Risks: <what could stall this>B. MEDDICC table with known / evidence / missing.
C. Handoff note for the AE: account, contact and role, problem confirmed (quote), business impact, urgency, decision process, current solution, other stakeholders, objections, commitments made, next step, unknowns, recommended follow-up. Each item tagged Said / Interpreted / Unknown.
D. Follow-up email to the prospect (under 120 words): thank them, recap their situation in their words, confirm the next step and date. No new pitch.
Human approval gate
The user reviews the CRM update and follow-up email before anything is logged or sent.
Quality check
- No field contains information that isn't in the notes.
- Next step has an owner and a date, or is flagged as missing.
- Follow-up email matches what was actually agreed.
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.

