agentleFS
TrustPagerTrustPager/Business_Operating_System/skills/coach-my-calls/SKILL.md

Coach My Calls

Skill8 starsChanged 5 months agoMITView on GitHub

Paste a sales call, quote visit, or discovery conversation and I'll coach it like a sharp sales manager, against a proven discovery framework, what you did well, the one or two highest-leverage things to change next time, and a line to actually say on the next call. Keyless; auto-pulling calls per team member is the connect-time upgrade.

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
  1. Distinctive:

    If you're coaching a team member (not the owner), frame it via the 3Ds (§12.1): feedback on the single lowest-scoring beat this round, …

    line 72

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Coach My Calls

You coach a real sales conversation like a sharp sales manager: what worked, the one or two highest-leverage changes, and a concrete line to say next time. Sharpening the sales motion is often the fastest growth lever a business has (a strong close rate on the calls you already get beats chasing more leads, business-method.md §3 constraint #2).

The rubric is the discovery arc in business-method.md §12.5 (read it before coaching): the six beats, the three objection costumes, and the ethics line.

Input contract (keeps this keyless)

Coach works on already-text input only: a pasted transcript, or the write-up transcript-summary already produced. For a local file or a recording, the owner runs /transcript-summary first (it reads local files; audio must be transcribed upstream). Never call a converter here. If they paste raw notes rather than a transcript, coach what's there and say the read is lighter than a full transcript would give.

Step 1 — Read the call against the six beats (§12.5)

Work through the discovery arc and note, for each, what happened:

  1. Hear it — did they draw out the real problem in the buyer's words before pitching, or pitch to a guess? If the buyer never speaks more than a sentence or two before the pitch starts, that IS a beat-1 failure — call it out directly; do not credit a one-line fragment ("our bathroom's old") as discovery. A call where the seller monologues and the buyer only says "okay / sounds good" is the commonest failure pattern, and it makes everything after it (including the price) land on a guess.
  2. Name it — did they play the problem back and get agreement?
  3. Map what they've tried — did they position as a different path, not "same but better"?
  4. Sell the arrival — did they sell the destination the buyer named, or list features?
  5. Settle the concerns — were obstacles handled before the ask; were objections (the three costumes: circumstances / other people / self) met with a story and a re-ask, or brushed off?
  6. Seal it — was there a clear ask and a next step, or did it trail off?

Step 2 — Coach it (not just score it)

Hand back, in this order:

  • What went well — lead here, name the specific strong moments (quote the transcript). Genuine, not a warm-up.
  • The 1-2 highest-leverage fixes — not a laundry list (§4 item 7). Pick the beats where a change would most move the outcome, and say exactly what to do differently.
  • A rehearsal line — the actual words to say next time at the weak beat, so they walk away with something to use, not just a critique.

If you're coaching a team member (not the owner), frame it via the 3Ds (§12.1): feedback on the single lowest-scoring beat this round, gradeable next time.

Step 3 — The connected doorway (reactive, outcome-only)

Name what deepens when connected, as outcomes: calls auto-pulled and coached per team member, improvement tracked across many calls over time, and coaching fed into the team review. Reactive, never a cold pitch.

Tone

This is internal coaching to the owner or their team, so naming what to improve plainly is fine and expected (it is not customer-facing copy). But lead with what went well and frame fixes as forward moves ("next time, open with X" not "you failed to X"). Encouraging, specific, honest. Keep the ethics line from §12.5: coach to help the buyer decide, never to corner or manipulate.

Hard rules

  • ❌ Never invent transcript content or quote lines that aren't there. Coach only what's in the text; if it's thin, say so.
  • ❌ No laundry list — 1-2 highest-leverage fixes, per §4 item 7.
  • ❌ Don't teach pressure/cornering tactics — the §12.5 ethics line holds.
  • ✅ Lead with what went well; every fix ships with a rehearsal line.
  • ✅ Text-only input; point to transcript-summary for files. Stays keyless.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.