GTM ICP Definition

Run an ICP definition or refinement workshop. Use when a team needs to define who their best customers are, align sales and marketing on targeting, identify...

Security Scan

Instruction-only workshop facilitator that is internally consistent with its stated purpose and requests no credentials or installs—low intrinsic risk, but be cautious about sharing sensitive customer data when using it.

Purpose & Capability

Name/description match the instructions: the SKILL.md describes a structured ICP workshop and the steps (best-customer analysis, churn analysis, segmentation, buying-committee mapping, FIRE criteria, and a template) align with that purpose. No unrelated binaries, env vars, or installs are requested.

Instruction Scope

Instructions remain within the workshop facilitator scope (ask users for customer examples, analyze patterns, produce a versioned ICP doc). It references another skill ('gtm-qualification-scoring') for the full FIRE rubric — that external dependency is reasonable but should be checked. The primary operational concern is data sensitivity: the skill expects users to supply customer/deal details, which may include confidential or personal information; the instructions do not instruct any file reads or network exfiltration, but platform behavior (storage/transmission) should be considered before pasting sensitive data.

Install Mechanism

No install spec and no code files — instruction-only. No downloads, packages, or binaries are required, so there is nothing being written to disk or executed by the skill itself.

Credentials

The skill declares no required environment variables, credentials, or config paths. There are no disproportionate access requests.

Persistence & Privilege

always is false and disable-model-invocation is false (normal platform default). The skill does not request persistent system-wide privileges or modify other skills' configs.

Assessment

This skill is coherent and low-risk technically, but it asks you to provide customer and deal details — avoid pasting PII, confidential revenue figures, or sensitive contract terms unless you have consent and understand where the platform stores/transmits inputs. Verify the referenced 'gtm-qualification-scoring' skill exists and is trustworthy before relying on it. If you plan to run workshops with real customer data, confirm platform privacy/storage policies or sanitize inputs (use anonymized examples) and keep a local copy/versioning of the generated ICP document.

Facilitate a structured ICP workshop. Output is a versioned ICP document teams can act on across sales, marketing, and growth.

Steps

1. Best customer analysis — ask for 3–5 best customers. Extract: what they share (industry, size, stack, GTM model), what made them buy (trigger, champion, pain), what outcome they got, why they stay.

2. Loss/churn analysis — ask about lost deals and churned accounts. Identify recurring patterns → defines Negative ICP.

3. Segment the market — group into 2–5 segments by: vertical, business model, size band, data/tech maturity, GTM motion fit (PLG vs sales-led).

4. Map the buying committee — per segment: Economic Buyer (approves budget), Technical Champion (evaluates), End User (daily use), Blocker (can kill deal).

5. Define FIRE criteria — translate each segment into measurable scoring signals. What firmographic attributes = high Fit? What behavioral signals = high Intent? Use the gtm-qualification-scoring skill for the full FIRE rubric.

6. Write the ICP document — use the template below.

Output (inline version)

ICP v[X] · [Company] · [Date]

Segment [#]: [Name]
Who: [1–2 sentences]
Firmographics: industry · size · geography · business model
Stack signals: [tools or tech patterns indicating fit]
Trigger events: [what makes them enter the market]
Buying committee: Economic Buyer / Champion / End User / Blocker
Why they buy: [pain + outcome]
ACV range / sales cycle: [estimate]

Negative ICP:
- [Characteristic] → [why it's a bad fit]

FIRE scoring criteria for this segment:
- Fit: [high-signal markers]
- Intent: [specific triggers]
- Recency: [timeframe threshold]
- Engagement: [interaction types that count]

Flag any segment not grounded in real customers as [Hypothesis — validate with first 10 customers]. Version the document and recommend revisiting every 6 months.