sales-call-scorer
Skill0 starsChanged 10 days agoMITView on GitHub
Scores a sales call 0–100 against the playbook ("did we run the play?") from a transcript. Use for "score this call", "grade my call", "how did that call go", or a transcript dropped for feedback. Scores execution (stage objectives, qualification depth, mechanics, next step), with an evidence quote for every sub-score. Qualification is scored on MEDDPICC. Evaluates only; never writes to the CRM. sales-intake calls it and does the writing.
In this repository12 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- call-debriefSkill
- company-contextSkill
- crm-recordsSkill
- humanize-copySkill
- meeting-prepSkill
- pipeline-analysisSkill
- product-contextSkill
- sales-intakeSkill
- sales-playbookSkill
- setupSkill
- weekly-exec-updateSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
Read the entire transcript: Never score from a summary.
≈ “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
Sales call scorer
Never writes the CRM. It returns a scorecard. Run it alone when someone only wants feedback.
Inputs
- A transcript, or a filled
sales-intake/INTAKE-FORM.md. - The deal's current stage, if a deal is matched (
crm-recordsmaps stage → call type).
Workflow
- Read the entire transcript. Never score from a summary. Confirm it is the right call: attendees, date, company.
- Determine the call type. From the deal stage, else from content. If the call matches an
earlier stage than the deal is in, score against the call's actual stage and raise
DEAL OVER-ADVANCED. If the call cleanly qualifies the prospect out, apply the qualified-out rule inRUBRIC.md. - Apply
RUBRIC.mdandPLAYBOOK.md. Four dimensions. Every sub-score cites a quote or records the explicit absence. Too little transcript to judge → "insufficient evidence", not a guess. - Return the scorecard.
Output
CALL SCORE: [N]/100 - Grade [A-F] ([call type], [date], [duration])
Deal: [name] [DEAL OVER-ADVANCED: call was X, deal is in Y]
A. Stage objectives ...... [n]/40
- [objective]: [MET/PARTIAL/MISSED] - "[quote]"
B. Qualification (MEDDPICC) [n]/25
- [letter] [signal] [n] - "[quote]" (out-of-scope letters: N/A)
MEDDPICC open: [letters still unknown across the deal log and this call, e.g. E, C-competition]
C. Call mechanics ........ [n]/20
D. Advancement ........... [n]/15
- Next step: "[quote or 'none secured']"
PLAYS: [RAN]/[PARTIAL]/[MISSED]/[N/A] per play in scope
STRENGTHS: top 3
GAPS: top 3 (each becomes a follow-up task)
DEEP BLOCKERS: ...
RECOMMENDED MOVE: one move
COACHING: two or three sentences, direct
SENTIMENT: on-track / at-risk / criticalPlain text, no emojis: the output is pasted into a CRM and Slack.
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.

