AGENT SKILL

gtm-sprint

Runs a 7, 14 or 21 day go-to-market (GTM) sprint: customer interviews, positioning, landing page and distribution, with go/no-go checkpoints. Use to test demand or launch an MVP.

License 许可
MIT
所属仓库
vibbs/company-os
最近更新
2026-10-21
Stars
★ 6
Forks
2

01 前言 Front Matter

name
gtm-sprint
description
Runs a 7, 14 or 21 day go-to-market (GTM) sprint: customer interviews, positioning, landing page and distribution, with go/no-go checkpoints. Use to test demand or launch an MVP.
argument-hint
[plan | status | gate-check] [7 | 14 | 21]
user-invocable
true

03 SKILL.md 正文

GTM Sprint

When to Use

  • Pre-product validation: "Do people actually want this?"
  • MVP launch: "We built it, now how do we get users?"
  • New market entry: "We're expanding to a new segment"
  • Repositioning: "Our current positioning isn't working"
  • Any time the user says: "go to market", "GTM", "launch plan", "validate market", "find customers"

Reference

  • ID: S-GRO-09
  • Category: Growth
  • Inputs: company.config.yaml, ICP artifact (or triggers creation), objective description
  • Outputs: GTM Sprint Plan artifact + GTM Sprint Report artifact -> artifacts/growth/
  • Used by: Growth Agent
  • Tool scripts: ./tools/artifact/validate.sh

Purpose

Orchestrates a time-boxed go-to-market sprint that sequences validation, commitment, launch, and distribution into a structured daily cadence with quantified decision gates. Unlike /ship (which builds software) or /rapid-prototype (which proves technical feasibility), /gtm-sprint validates market demand and builds distribution — no code required.

Inspired by the 100x GTM Playbook methodology: validate with interviews, secure commitments (LOIs), build launch assets, activate distribution channels, and measure traction — all within a fixed sprint window.

When NOT to Use

  • Building software features -> use /ship
  • Technical proof-of-concept -> use /rapid-prototype
  • Single-channel optimization -> use /channel-playbook directly
  • Post-launch growth optimization -> use individual growth skills directly

What You Will See

/gtm-sprint plan 14 produces a plan like this (excerpt). Each day lists tasks, the skill to use and the output.

GTM Sprint Plan: Acme Invoices, 14 days, stage: idea

Day 0  Prep: lock ICP and outreach
  - Finalize ICP (role, company size, budget, current workaround)
  - Build a 100-name prospect list; book 15+ calls for Days 1-2
  - Skill: icp-positioning        Output: ICP artifact, prospect list
Day 1  Pain discovery: Mom-Test interviews (10+, 15 min each)
  - Skill: customer-conversations prep mom-test
  - Output: 10+ CONV- artifacts
Day 2  Insights: synthesize pain data
  - Skill: feedback-synthesizer   Output: Pain Matrix
  Gate: Interview Volume (>= 15 completed)

/gtm-sprint gate-check evaluates one gate and tells you what to do next:

Gate: Interview Volume (Day 2)   Metric: completed interviews
Threshold: >= 15   Actual: 11   Result: FAIL (advisory at stage idea)
Corrective action: extend Days 1-2 by 1-2 days and increase outreach volume.
Adjust the plan, override the gate, or pause the sprint?

Answer with adjust, override or pause. Gates are advisory at idea stage and enforced at growth and scale.

Modes

Mode: plan (default)

Generates a personalized GTM Sprint Plan with daily tasks, decision gates, and skill sequencing.

Mode: status

Reviews progress against the active GTM Sprint Plan — checks completed tasks, gate metrics, and recommends adjustments.

Mode: gate-check

Evaluates a specific decision gate and recommends corrective action if thresholds are not met.

Procedure

Phase 0: Context & Configuration

  1. Load Context

    • Read company.config.yaml for company name, stage, product description
    • Check artifacts/product/ for existing ICP artifact
    • Check artifacts/decision-memos/ for existing CONV- artifacts (interview count)
    • Check artifacts/growth/ for existing positioning, landing page, channel playbooks
  2. Sprint Configuration

    Ask the user for (with smart defaults):

    Parameter Default Options
    Sprint duration 14 days 7 / 14 / 21 days
    Start date Today Any date
    Objective From product description Custom
    Target LOIs 5 3-10
    Target interviews 20 10-30
    Target wait-list 50 25-200

    Stage-aware defaults:

    • idea: 14 days, 20 interviews, 5 LOIs, 50 wait-list
    • mvp: 7 days (validation done), 10 interviews, 3 LOIs, 100 wait-list
    • growth: 7 days (repositioning sprint), 15 interviews, 5 LOIs, 200 wait-list
  3. ICP Check

    • If no ICP artifact exists: inform the user — "No ICP found. The sprint will begin with ICP definition (Day 0). Run /icp-positioning now, or we'll include it in the sprint plan."
    • If ICP exists but is draft: note it needs validation through interviews

Phase 1: Generate Sprint Plan (Mode: plan)

  1. Build Daily Task Schedule

    Generate a day-by-day plan adapted to the sprint duration. For a 14-day sprint:

    Week 1: Validate & Commit (Days 0-6)

    Day 0 — Prep: Lock ICP & Outreach

    • Finalize ICP criteria (role, company size, budget, current workaround)
    • Build prospect list (100 targets) — use LinkedIn, communities, competitor reviews
    • Book 15+ discovery calls for Days 1-2
    • Skill: icp-positioning (if no ICP exists)
    • Skill: customer-conversations prep cold-outreach
    • Output: ICP artifact (draft or updated), prospect list

    Day 1 — Pain Discovery: Mom-Test Interviews

    • Conduct 10+ Mom-Test interviews (15 min each)
    • Record key quotes and $$ impact of each pain point
    • Focus: current behavior, cost of status quo, workarounds tried
    • Skill: customer-conversations prep mom-test
    • Skill: customer-conversations debrief (after each call)
    • Output: 10+ CONV- artifacts

    Day 2 — Insights: Synthesize Pain Data

    • Conduct remaining interviews (target: 20 total)
    • Run feedback synthesis on all CONV- artifacts
    • Produce Pain Matrix: themes clustered by $$ impact + urgency
    • Rank themes, identify top 3 "hair-on-fire" problems
    • Skill: feedback-synthesizer (with pain-matrix output)
    • Output: Feedback report with Pain Matrix

    Day 3 — Refine: Sharpen MVP Scope

    • Map current workflow -> desired workflow for top pain
    • Update MVP scope based on Day 2 insights
    • Quantify time saved and cost saved per user
    • Draft value proposition: "We help X do Y, saving Z hours/week"
    • Skill: prd-writer or update existing PRD
    • Output: Updated PRD or mini-PRD (if pre-product)

    Day 4 — Commit: Secure Letters of Intent

    • Draft 1-page LOI template (problem, proposed solution, commitment level, timeline)
    • Re-contact top 5 prospects from interviews
    • Show prototype/mockup/demo and request commitment
    • Obtain verbal commitments or e-signatures
    • Output: LOI artifacts (target: 5 signed)

    Day 5 — Narrative: Go Public

    • Write 1-sentence mission statement
    • Write 2-paragraph vision statement
    • Craft positioning: "We help [X] do [Y] in [Z] seconds"
    • Publish first build-in-public post
    • Skill: positioning-messaging
    • Skill: channel-playbook (build-in-public arc)
    • Output: Positioning artifact, first public post draft

    Day 6 — Gate Check: Week 1 Review

    • Review metrics: interviews completed, LOIs signed, wait-list size
    • Decision Gate 1: Evaluate against thresholds (see Decision Gates below)
    • Adjust Week 2 plan based on gate results
    • Output: Week 1 status update

    Week 2: Launch & Grow (Days 7-13)

    Day 7 — Launch Pad: Landing Page

    • Build landing page with top 3 pain quotes above the fold
    • Embed email capture / wait-list form
    • Add Calendly or booking link for demos
    • Skill: landing-page-copy
    • Output: Landing page copy artifact (or deployed page)

    Day 8 — Lead Magnet: Field Report

    • Compile interview insights into a downloadable asset
    • Format: "Field Report" PDF, industry brief, or data snapshot
    • Gate the download behind email capture
    • Skill: content-engine (lead-magnet format)
    • Output: Lead magnet content artifact

    Day 9 — Distribution: Directory Blitz

    • Submit product to 15+ relevant directories
    • Craft directory-specific hooks using interview data
    • Skill: channel-playbook (directory-blitz template)
    • Output: Directory submission tracker

    Day 10 — Community: Engagement Strike

    • Join 3-5 Slack/Discord communities where ICP hangs out
    • Share one fresh insight chart from interview data
    • Offer free office-hours or AMA slot
    • Decision Gate 2: Check wait-list size against threshold
    • Skill: channel-playbook (community tactics)
    • Output: Community post drafts

    Day 11 — Competitor: Poaching Play

    • Identify 50-100 competitor reviewers (G2, Capterra, Reddit)
    • Send value-first outreach: "What's frustrating about [competitor]?"
    • Offer concierge migration or white-glove onboarding
    • Skill: channel-playbook (competitor-poaching template)
    • Output: Competitor outreach templates

    Day 12 — Demo Day: Show & Tell

    • Live demo to LOI signers and top prospects
    • Capture video testimonials (even 30-second clips)
    • Post build-in-public recap with traction metrics
    • Output: Demo script, testimonial clips, recap post draft

    Day 13 — Sprint Retrospective

    • Review all metrics vs. targets
    • Decide top 3 growth experiments for next 30 days
    • Publish internal retro + public "what's next" post
    • Skill: retrospective (or standalone retro)
    • Output: GTM Sprint Report artifact

    Sprint Duration Variants

    7-Day Sprint (compressed):

    • Days 0-1: ICP + Interviews (combine Prep + Pain Discovery)
    • Day 2: Synthesize + Refine MVP
    • Day 3: LOIs + Positioning
    • Day 4: Landing Page + Lead Magnet
    • Day 5: Distribution Blitz + Community
    • Day 6: Demo Day + Retrospective
    • Gate Check: Day 3 (mid-sprint)

    21-Day Sprint (extended):

    • Week 1: Days 0-6 (same as 14-day)
    • Week 2: Days 7-13 (same as 14-day)
    • Week 3: Days 14-20 — Optimize & Scale
      • Day 14-15: A/B test landing page variants
      • Day 16-17: Double down on best-performing channel
      • Day 18: Pitch deck assembly
      • Day 19: Investor/partner outreach
      • Day 20: Sprint close + 30-day plan
    • Gate Check: Day 6, Day 13, Day 18
  2. Adapt for Existing Progress

    Skip steps where artifacts already exist:

    • ICP artifact exists and is approved -> skip Day 0 ICP work
    • 10+ CONV- artifacts exist -> reduce Day 1-2 interview targets
    • Positioning artifact exists -> skip Day 5 positioning (just do build-in-public)
    • Landing page copy exists -> skip Day 7 copy work (focus on deployment)

Phase 2: Decision Gates

  1. Define Quantified Gates

    Each gate has a metric, threshold, and corrective action:

    Gate Day Metric Threshold If Below
    Interview Volume 2 Completed interviews >= 15 (of 20 target) Extend Days 1-2 by 1-2 days, increase outreach volume
    Pain Clarity 3 Distinct pain themes with $$ quantification >= 3 themes Broaden interview scope to adjacent ICP segments
    LOI Commitment 6 Signed/verbal LOIs >= 60% of target Double interview volume in Week 2, revisit ICP criteria
    Wait-list Traction 10 Wait-list sign-ups >= 50% of target Rerun positioning, test different channels, check landing page conversion
    Distribution Reach 12 Directory submissions + community posts >= 10 combined Focus remaining days on single highest-ROI channel

    Gate Severity by Stage:

    • idea: All gates are advisory (warnings) — learning is the goal
    • mvp: LOI and Wait-list gates are enforced — market signal is critical
    • growth/scale: All gates enforced — must hit thresholds to continue
  2. Gate Check Procedure (Mode: gate-check)

    When a gate is triggered:

    1. Count the relevant metric from artifacts (CONV- count, LOI count, etc.)
    2. Compare against threshold
    3. If PASS: congratulate, move to next phase
    4. If FAIL: present corrective action with specific steps
    5. Ask user: "Adjust the plan, override the gate, or pause the sprint?"

Phase 3: Sprint Artifacts

  1. GTM Sprint Plan Artifact

    Save to artifacts/growth/GTM-SPRINT-{date}.md:

    ---
    id: GTM-SPRINT-{date}
    type: launch-brief
    title: "GTM Sprint Plan — {objective}"
    status: draft
    created: {date}
    author: Growth Agent
    sprint_duration: 14
    sprint_start: {date}
    sprint_end: {date+duration}
    target_interviews: 20
    target_lois: 5
    target_waitlist: 50
    ---
    

    Body contains the full daily task schedule with checkboxes.

  2. GTM Sprint Report Artifact (end of sprint)

    Save to artifacts/growth/GTM-SPRINT-REPORT-{date}.md:

    ---
    id: GTM-SPRINT-REPORT-{date}
    type: launch-brief
    title: "GTM Sprint Report — {objective}"
    status: review
    created: {sprint_end_date}
    author: Growth Agent
    parent: GTM-SPRINT-{date}
    ---
    

    Body contains:

    • Metrics vs. Targets: interviews, LOIs, wait-list, directory submissions, community engagement
    • Top Learnings: 3-5 key insights from interviews
    • Pain Matrix Summary: top 3 validated pain points with $$ impact
    • Positioning Validation: what resonated, what didn't
    • Channel Performance: which channels drove most traction
    • Next 30 Days: top 3 growth experiments recommended
    • Decision: proceed to /ship, pivot positioning, expand ICP, or pause

Phase 4: Status Tracking (Mode: status)

  1. Progress Dashboard

    When invoked with status:

    1. Find the most recent GTM-SPRINT- artifact in artifacts/growth/
    2. Calculate current sprint day based on sprint_start date
    3. Count completed metrics:
      • CONV- artifacts created since sprint start
      • LOI artifacts in artifacts/product/
      • Landing page / positioning artifacts created
      • Channel playbook artifacts created
    4. Present progress:
    ## GTM Sprint Status — Day {N} of {duration}
    
    ### Metrics
    | Metric | Target | Current | Status |
    |--------|--------|---------|--------|
    | Interviews | 20 | 12 | On track |
    | LOIs | 5 | 2 | At risk |
    | Wait-list | 50 | 8 | Behind |
    
    ### Today's Focus
    - [Day N tasks from the plan]
    
    ### Upcoming Gate
    - Gate: LOI Commitment (Day 6) — need 3 more LOIs in 2 days
    

LOI (Letter of Intent) Template

When creating LOI artifacts on Day 4, save to artifacts/product/LOI-{company}-{date}.md using this structure:

---
id: LOI-{company}-{date}
type: loi
title: "Letter of Intent — {Company Name}"
status: draft
created: {date}
author: {user}
parent: {icp-artifact-id}
---

LOI Body Template

## Letter of Intent

**From**: {Prospect Name}, {Role} at {Company}
**To**: {Your Company}
**Date**: {date}

### Problem Acknowledged
{1-2 sentences describing the pain point this prospect validated}

### Proposed Solution
{1-2 sentences describing what you plan to build/offer}

### Commitment Level
- [ ] Verbal interest
- [ ] Email confirmation
- [ ] Signed LOI
- [ ] Pilot agreement with timeline

### Terms (if applicable)
- **Pilot Duration**: {e.g., 30 days}
- **Pilot Scope**: {what they'll use}
- **Success Criteria**: {what "working" looks like for them}
- **Pricing Discussion**: {willingness-to-pay signal}

### Prospect Context
- **Company Size**: {employees / revenue range}
- **Current Workaround**: {how they solve this today}
- **$$ Impact**: {estimated cost of current pain per month/year}
- **ICP Fit**: Strong / Partial / Weak
- **Decision Timeline**: {when they need a solution}

### Next Steps
{Agreed follow-up actions}

Cross-References

  • icp-positioning (S-PROD-01) — Day 0 ICP definition
  • customer-conversations (S-PROD-08) — Days 1-2 interview prep/debrief, Mom-Test mode
  • feedback-synthesizer (S-PROD-04) — Day 2 pain synthesis with Pain Matrix
  • prd-writer (S-PROD-02) — Day 3 MVP refinement
  • positioning-messaging (S-GRO-01) — Day 5 narrative and positioning
  • landing-page-copy (S-GRO-02) — Day 7 launch page
  • content-engine (S-GRO-07) — Day 8 lead magnet
  • channel-playbook (S-GRO-04) — Days 9-11 distribution (directory blitz, community, competitor poaching)
  • retrospective (S-ORG-14) — Day 13 sprint retro
  • ship (S-ORG-08) — post-sprint graduation to full build flow

Quality Checklist

  • [ ] Sprint duration is appropriate for company stage
  • [ ] Daily tasks are specific and actionable
  • [ ] Decision gates have quantified thresholds
  • [ ] Existing artifacts are accounted for (no duplicate work)
  • [ ] Skill sequencing follows dependency order
  • [ ] LOI template is included for Day 4
  • [ ] Sprint plan adapts to 7/14/21 day variants
  • [ ] Sprint report captures metrics vs. targets
  • [ ] Corrective actions are specific for each failed gate
  • [ ] Artifact passes validation

04 同仓库更多 Skills