Call Coaching Scan

• Source: Built by Jonathan Kvarfordt on the GTM AI OS. Supplied as a complete zip and served unchanged.
• Frameworks applied: Role scans

It covers a job one of the eight revenue seats has to finish, and it states its process, decision rules, and output format clearly enough for an agent to follow.

1. 1.Copy the SKILL.md text below, or download the raw file.
2. 2.Create a folder named exactly call-coaching-scan in your agent's skills directory.
3. 3.Save the file inside that folder as SKILL.md.
4. 4.Ask the agent one of the trigger requests below.
5. 5.Check the output against what you already know before it leaves your desk.

• how did I do on this call
• coach me
• review my discovery call

---
name: call-coaching-scan
description: 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. Use when someone asks "how did I do on this call", "coach me", "review my discovery call", "give my rep feedback", "call review", or "what should I have said".
---

# Call Coaching Scan

For any rep, their manager, or enablement. Grades the seller only, never the buyer. Every point ties to a moment in the call and comes with words to use next time.

Rebuilt from the Momentum prompt library (Discovery Call Review, Demo Call Review, Pricing Proposal Call Review, Negotiation Call Review, Analyze an Entire Opportunity, MEDDICC Call Review coaching format).

## Inputs

- The raw transcript (this skill needs seller lines, which the ledger drops) and `ledger.md` for buyer-side facts.
- `rep-read.md` if present.
- Who is the seller: from `gtm-context.md` (company name or email domain) or ask. Grade only seller-side speakers.
- Call stage: ask if it is not obvious from the meeting title. One of: discovery, demo, pricing or proposal, negotiation, CS check-in.

## Step 1: Count before judging

| Metric | How | Benchmark |
|---|---|---|
| Talk-to-listen | seller words / total words | discovery 30/70 seller/buyer; demo 50/50; negotiation 40/60 |
| Longest seller monologue | longest uninterrupted seller turn, in words or seconds | under 90 seconds |
| Questions asked | count seller questions, split open vs closed | discovery: 10+ open |
| Pains captured vs addressed | PAIN rows in ledger; how many the seller asked a follow-up on | every PAIN gets a follow-up |
| Next step | dated COMMITMENT row exists | yes |

If the transcript has no reliable speaker labels, say the ratio is approximate.

## Step 2: Score the stage rubric, 0 to 4

0 not done, 1 barely, 2 somewhat, 3 well, 4 thoroughly.

| Stage | Criteria |
|---|---|
| Discovery | situation questions first; pain uncovered; impact quantified; decision process and timeline; budget; active listening (summarize and confirm); no early pitching; next step |
| Demo | tied to stated pains; tailored to the buyer's workflow; checked for reactions; handled objections; technical credibility; next step |
| Pricing or proposal | price stated clearly and confidently; value tied to price; pricing objections handled; concessions traded, not given; commitment asked for |
| Negotiation | opening and anchoring; concession management; value defended; objections handled; clear give-get; closing and next step |
| CS check-in | outcomes reviewed against goals; adoption and usage; risks surfaced; expansion explored; next step |

Each score cites a timestamp or says "not observed".

## Step 3: Coded moments

| Time | Category | Excerpt (seller or buyer) | What happened | Coaching note |

Pick the 5 to 10 moments that matter most: a missed follow-up on a pain, a pitch too early, a great question, an objection, the close.

## Step 4: Coaching

- **Top 3 strengths**, each with the excerpt that shows it.
- **Top 3 fixes**, each with: the excerpt, what to say instead (exact words), and why it works.

Alternate wording rules: open-ended (How, What, Walk me through), funnel from broad to specific, summarize and confirm, quantify impact, never leading.

Example fix:
- Moment (14:22): buyer says "our forecast is mostly vibes". Seller moves to the demo.
- Say instead: "When you say mostly vibes, what happened the last time the forecast missed? What did that cost you?"
- Why: the buyer handed you a pain with emotion. A follow-up turns it into impact you can quote later.

## Step 5: Rules for the next call

Three dos and three don'ts, specific to this rep and this deal.

## Output: `coaching.md`

```markdown
# Call Coaching: <rep>, <account>, <stage>, <date>
Rubric score: <sum>/<max> | Talk-to-listen: <x>/<y> | Open questions: <n> | Pains followed up: <a> of <b>

## Numbers
## Rubric
| Criterion | Score | Evidence (time) |
## Coded moments
## Strengths
## Fixes (say this instead)
## Next call: do / don't
## One thing to practice this week
```

## Rules

- Grade seller-side speakers only.
- Every claim cites a timestamp or excerpt. No generic sales advice.
- Top 3, not top 10. A rep can act on three things.
- If the Rep Read shows the rep noticed something live (a `D:` or `C` reason), credit it. Noticing is a skill.

What does the Call Coaching Scan skill do?

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.

Where does the Call Coaching Scan skill come from?

Built by Jonathan Kvarfordt on the GTM AI OS. Supplied as a complete zip and served unchanged.. The source and included files are shown on this page and in the package download.

Why was the Call Coaching Scan skill chosen for this library?

It covers a job one of the eight revenue seats has to finish, and it states its process, decision rules, and output format clearly enough for an agent to follow.

When should the Call Coaching Scan skill not be used?

Use it only for the job named above. A skill applied outside its scope produces confident output on the wrong question.

How do I install the Call Coaching Scan SKILL.md file?

Download the file, create a folder named exactly call-coaching-scan inside your agent's skills directory, and save the file inside it as SKILL.md. The agent loads it when a request matches the description.