zime-gtm-skills

If you're building call-review or deal-inspection tooling in-house, start
here. 16 Agent Skills that audit sales call
transcripts and CRM exports against GTM rubrics: one file in, one
evidence-cited audit out. Runs locally, no credentials, no data leaves your
machine. Works with Claude Code, Cursor, Windsurf, OpenAI Codex, or any agent
that reads the Agent Skills spec.

Two axes: deal stage (discovery through renewal) and initiative
(MEDDICC, BANT, pain identification: frameworks that run at any stage).

The rule every skill here is built on: an uncited finding doesn't ship.
Every dimension a skill marks Covered, Partial, or Missed points at a direct
quote or timestamp from the transcript in front of it, never a plausible
guess dressed up as a finding.

MIT, no gating, no telemetry, no lead capture. Fork it, rewrite the rubrics
for your motion, ship it internally: that's the intended use. Built and
maintained by Zime; we'd rather hand over the rubrics than
argue anyone out of building.

Contributions welcome: see CONTRIBUTING.md. Found a
bug or have a question? Open an issue.

• See it work
• Quick install
• How these fit together
• Coverage: two dimensions, not one
• Where this stops
• Available skills
• Repository structure
• FAQ
• Installation
• Built by Zime
• License

See it work

Same synthetic call, same model, two ways: a plain "review this call"
prompt with no skill, and deep-discovery against
skills/deep-discovery/assets/sample-transcript.txt.

Without the skill: unaided, prompted "review this discovery call"

Discovery call feedback — Northwind x Vendor Co:

Strong:
- Quantified pain: 8-10 hrs/week dispatcher time chasing exceptions.
- Got compelling event: lost account last quarter + board-mandated
  ops review this quarter.
- Got economic buyer + threshold: CFO approves anything over $40k/year.

Gaps:
1. No next step. Call ends on rapport with zero concrete action.
2. Budget not probed directly — got the approval threshold, not the
   actual budget range.
3. Timeline vague — "before Q3 planning" never pinned to a date.

No structure that repeats across calls, no consistent evidence format,
findings and gaps interleaved with commentary.

With deep-discovery

1. Pain & business impact — Covered
   "spends probably 8-10 hours a week just chasing down exception
   statuses"

2. Current state & tooling — Covered
   "Dispatchers log exceptions in a shared spreadsheet"

3. Trigger — why now — Covered
   "we lost a mid-size account last quarter over a missed exception,
   and our board asked for a logistics-ops review this quarter"

[... 4. Desired outcome, 5. Decision process — both Covered, trimmed ...]

6. Budget & timeline — Partial
   CFO approval threshold ($40k) surfaced, but that's an approval
   gate, not deal-specific budget. No explicit decision-date beyond
   "before Q3 planning."

7. Procurement & buying path — Covered
   "anything touching our TMS goes through a security review with
   our IT director. Usually takes about three weeks"

[... 8. Competitive landscape, 9. Rapport — both Covered, trimmed ...]

Reads-well-too check: 8/9 Covered, 1 Partial, zero Missed — matches a
call that clearly covered discovery well. No fault-finding bias here.

Highest-leverage next steps:
1. Nail a real budget number, not just the approval gate.
2. Pin a hard decision date, not "before Q3 planning."

Same nine dimensions, every run: a rep gets the same read whoever's on
the call, and every Missed or Partial is falsifiable against the transcript
next to it.

Run it yourself, nothing of your own required:

git clone https://github.com/zime-ai/zime-gtm-skills.git
cd zime-gtm-skills
claude "run deep-discovery on skills/deep-discovery/assets/sample-transcript.txt"

Quick install

# Claude Code plugin (recommended)
/plugin marketplace add zime-ai/zime-gtm-skills
/plugin install gtm-skills@zime-gtm-skills

# or clone and copy into any agent that reads the Agent Skills format
git clone https://github.com/zime-ai/zime-gtm-skills.git
cp -r zime-gtm-skills/skills/* .agents/skills/

See Installation below for the submodule option and the
.claude/skills/ vs .agents/skills/ distinction.

How these fit together

16 skills: 11 stage motions laid out across the deal lifecycle and 5
cross-stage initiative skills.

%%{init: {"flowchart": {"nodeSpacing": 65, "rankSpacing": 90, "padding": 20}, "themeVariables": {"fontSize": "20px"}}}%%
flowchart LR
    subgraph nb["New business"]
        direction LR
        dd[deep-discovery] --> mq[meeting-to-qualify]
        mq --> td[technical-discovery]
        td --> id[improve-demo]
        id --> pc[pilot-to-conversion]
        pc --> nc[negotiation-closing]
    end
    subgraph ps["Post-sale"]
        direction LR
        oj[onboarding-journey] --> cs[customer-success]
    end
    nb --> ps
    fcr["first-call-rampup, coaching at any stage"]
    cp["churn-prevention, renewal risk"]
    ue["upsell-expansion, expansion opportunity"]
    ps -.-> cp
    ps -.-> ue

Unlike some agent-skill collections, there's no foundational context skill
every other one depends on, and no required order. Each skill is independent:
one call transcript (or CRM export) in, one evidence-cited audit out.

Coverage: two dimensions, not one

Stage and initiative are both skills: a flat set, no nesting.

	Stage	Initiative
deep-discovery	Discovery	n/a
meeting-to-qualify	Qualify	n/a
technical-discovery	Technical discovery	n/a
improve-demo	Demo	n/a
pilot-to-conversion	Pilot	n/a
negotiation-closing	Negotiation	n/a
first-call-rampup	Any (coaching)	n/a
onboarding-journey	Onboarding	n/a
customer-success	Post-sale review	n/a
churn-prevention	Renewal risk	n/a
upsell-expansion	Expansion	n/a
meddicc	Any	MEDDICC
bant	Any (early)	BANT
pain-finder	Any	Identify pain
next-step-commitment	Any	Next step commitment
adoption-leaderboard	Any (post-sale)	Behavior adoption

Roadmap. The named methodologies (MEDDPICC, FAINT, Sandler, Challenger,
Sell the Dream, and others) aren't built yet. See MAINTAINING.md's
deferred work, or contribute one.

Where this stops

Everything here runs on one file, in one session, with no memory. That's a
design ceiling, and it's worth knowing before you build on it:

• One call at a time. You point a skill at a transcript. Auditing every
  call an org makes, as it happens, is a pipeline problem, not a rubric
  problem.
• Snapshot, not trend. Each run is independent. Nothing here tracks the
  same rubric across a rep's quarter, a team, or a pipeline over time.
• Reads exports, never writes back. A CRM .csv goes in. Nothing updates
  a deal record or feeds a forecast.
• Rubrics are hand-written, not calibrated. These dimensions come from
  experience and community correction, not from regression against your own
  won/lost data. A sound default, not a model fit to your motion.
• No rep or manager surface. This is an engineer running a command.
  Assignment, coaching loops, adoption tracking are out of scope.

Past that ceiling you're building a platform, not a skill, and that's a much
larger piece of work. It's the problem Zime works on.

Available skills

16 skills across new business, post-sale, and cross-stage initiative

New business

Skill	Audits	Input
deep-discovery	Early-stage discovery call quality — pain, trigger, decision process, budget/timeline, procurement, competition, rapport	transcript (.txt/.vtt/.json/.md) or CRM export (.csv)
meeting-to-qualify	Whether a deal is real — authority, budget plausibility, urgency, fit, competitive situation, concrete next step	transcript or CRM export
technical-discovery	Whether a POC is set up to succeed — current stack, future-state fit, scope clarity, technical stakeholders, resourcing	transcript or CRM export
improve-demo	Demo quality — tailoring, stakeholder presence, engagement, competitive differentiation, clear next step	transcript or CRM export
pilot-to-conversion	Pilot/POC health — plan alignment, technical readiness, user engagement, results against criteria, conversion path	transcript or CRM export
negotiation-closing	Late-stage deal momentum — pricing reaction, procurement progress, contracting awareness, closure timeline	transcript or CRM export
first-call-rampup	New-rep foundational technique — opening, open-ended questioning, pain uncovering, pitch clarity, rapport	transcript only

Post-sale

Skill	Audits	Input
onboarding-journey	New-customer onboarding health — setup experience, sentiment, engagement, milestone clarity	transcript or CRM export
customer-success	Periodic QBR/health-review depth — value vs. original goals, usage breadth, issue resolution, strategic alignment	transcript or CRM export
churn-prevention	Renewal risk — value realization, usage depth, champion strength, sentiment, service friction	transcript or CRM export
upsell-expansion	Expansion readiness — account health check first, opportunity clarity, champion strength, strategic alignment	transcript or CRM export

Initiative (cross-stage)

Skill	Audits	Input
meddicc	MEDDICC coverage — Metrics, Economic buyer, Decision criteria, Decision process, Identify pain, Champion, Competition	transcript or CRM export
bant	BANT coverage — Budget, Authority, Need, Timeline — lighter-weight, early-stage qualification	transcript or CRM export
pain-finder	Ranked pain points with confidence and evidence	transcript
next-step-commitment	Whether a call ended in a real, dated, two-sided commitment vs. a vague "we'll circle back"	transcript or CRM export
adoption-leaderboard	Ranks reps by adoption of five fixed winning behaviors (rapport, upsell signals, renewal risk, customer experience, value realization) across recent calls, lowest first	transcript or connector

All 11 stage motions are covered; the initiative set is five skills today.
New skills welcome, see CONTRIBUTING.md. For how these
are tested, and what that testing can and can't prove, see
EVALS.md.

Repository structure

zime-gtm-skills/
├── .claude-plugin/
│   ├── marketplace.json       # /plugin marketplace add zime-ai/zime-gtm-skills
│   └── plugin.json
├── skills/
│   └── skill-name/
│       ├── SKILL.md          # required
│       ├── references/       # optional, the rubric lives here
│       ├── assets/           # synthetic sample transcript/CSV
│       └── evals/            # declarative evals.json
├── validate-skills.sh         # local, zero-dep frontmatter/layout check
├── scripts/                    # check-docs-sync.sh, scan-content.py, pr-prep.sh
├── tests/                      # fixture tests for the validators above
├── evals/                       # eval methodology assets, see EVALS.md
├── CONTRIBUTING.md
├── MAINTAINING.md              # process + decisions, for picking this repo back up
├── LICENSE
└── README.md

FAQ

Do I need an API key or any credentials? No. Every skill is a markdown
rubric your agent reads and applies to a file you already have. Nothing
here calls out to a service, and nothing you feed a skill leaves your
machine beyond whatever your agent's own model call already sends.

My agent isn't picking up the skill. Claude Code reads
.claude/skills/, not .agents/skills/ — if you installed via Option 2 or
3 below for Claude Code specifically, copy (or symlink) into .claude/skills/
instead. For other agents, confirm they read the
Agent Skills spec format and that the skill
directory is where that agent expects it.

A skill's output doesn't look right. If it's the rubric itself — a
missed dimension, a wrong classification — that's the most useful kind of
issue, see CONTRIBUTING.md. If it's install
or tooling breaking, open a bug report.

Installation

Option 1: Claude Code plugin

/plugin marketplace add zime-ai/zime-gtm-skills
/plugin install gtm-skills@zime-gtm-skills

Option 2: Clone and copy

git clone https://github.com/zime-ai/zime-gtm-skills.git
cp -r zime-gtm-skills/skills/* .agents/skills/

Or copy just the one skill you want: cp -r zime-gtm-skills/skills/deep-discovery .agents/skills/.

[!TIP]
Claude Code reads .claude/skills/, not .agents/skills/ — see the
FAQ if your agent isn't picking the skill up.

Option 3: Git submodule

git submodule add https://github.com/zime-ai/zime-gtm-skills.git .agents/zime-gtm-skills

Update later with git submodule update --remote.

Built by Zime

Zime builds sales-call analytics; this repo is the
rubric layer we'd rather hand over than gatekeep. If it's useful, a star
helps other GTM/RevOps teams find it. Business inquiries: [email protected].

License

MIT, see LICENSE.

Maintained by Zime.