Product Context
Save ICP, voice, and scoring once so other skills stop asking
$ npx skills add sidchaudhary/gtm-skills/skills/product-contextWhat it does
Captures your ICP, brand voice, lifecycle stages, scoring model, and design preferences in one place.
You'll know it's time when...
Skills keep asking for the same setup details, and outputs are drifting off-brand.
How it works
Run it in three steps
Install
Copy the install command above and run it in your project.
Ask Claude
Ask for what you need in plain English, no prompt tuning required.
Get the output
Claude returns a structured artifact aligned to your ICP and voice.
Context
- This skill CREATES
.agents/product-context.md. It does not depend on an existing context file. - If
.agents/product-context.mdalready exists, read it, show the user its current Version and the last 2-3 Changelog entries so they see where the doc stands, then ask whether to update it or start fresh.
Inputs
-
Ask the following questions ONE AT A TIME. Wait for each answer before asking the next. If the user provides multiple answers at once, accept them and only ask remaining questions.
a. Company: What is your company name, website URL, and a one-line description of what you do?
b. Business model: Which best describes your model: SaaS, eCommerce, marketplace, or other?
c. ICP: Who is your ideal customer? (persona/title, company size, industry)
c2. Pain points: What are their top 2-3 pain points?
d. Buying committee: Who else is involved in the buying decision? List each role and their primary concern. If single decision-maker or B2C, note that and skip buying committee mapping.
e. Brand voice: Pick a voice: warm, bold, playful, authoritative, or direct. Are there any banned words or required themes?
f. Customer lifecycle: Use the default 6 stages (At Risk, Needs Attention, New, Promising, Regulars, Champions) or define custom stages? If the user picks defaults, read
references/lifecycle-stages.mdfor default stage definitions and scoring weights.g. Scoring: What does "high intent" look like for your business? What does "at risk" look like?
h. Design: What is your primary brand color? Style preference: minimal, editorial, bold, or luxury? Any preferred creative angles or imagery?
i. Metrics: What is your north star metric? List 2-3 secondary metrics and current baselines if known.
j. Competitive landscape: Who are your 2-3 main competitors, and for each, the one-line reason a prospect picks you over them (or them over you)? If the user doesn't know, note it as [NEEDS INPUT] rather than guessing a competitor's positioning.
k. Objections: What are the top 2-3 objections you hear in sales or support, and how do you actually respond to each? If none are known yet, note [NEEDS INPUT].
l. Switching dynamics: What pushes a customer to finally switch away from their current solution (or from doing nothing), and what makes them anxious about switching to you? One or two sentences each is enough.
Process
-
Before assembling, confirm your understanding: "Here's what I captured. Let me know if anything needs correcting." Show a brief summary of each section. Wait for confirmation or corrections.
-
After confirmation, assemble a structured markdown document with these sections. If any section has incomplete information, flag it as [NEEDS INPUT] rather than guessing.
- Company: name, URL, description, business model
- ICP: persona, company size, industry, pain points
- Buying Committee: table with columns: Role | Primary Concern
- Brand Voice: tone, banned words, required themes
- Lifecycle Stages: table with columns: Stage | Definition | Key Signals
- Scoring: high-intent definition, at-risk definition
- Design: primary color, style, creative angles
- Metrics: north star, secondary metrics, baselines
- Competitive Landscape: table with columns: Competitor | Why They Win / Why You Win
- Objections: table with columns: Objection | Response
- Switching Dynamics: what pushes them to switch, what makes them anxious about switching
-
Set the version and changelog. This is the paper trail every other gtm skill reads:
- New document: set
Version: v1and a single Changelog entry:- v1 (today's date): Initial context. - Updating an existing document: increment the version (v1 → v2), and prepend a new Changelog entry at the top of the list (newest first) summarizing what changed and why in one line, e.g.
- v3 (2026-08-01): Updated ICP after 5 customer interviews; added competitor Acme.Never rewrite or reorder past entries. - Pure typo-only fix: don't bump the version or add an entry. Just save the correction. Any real content change bumps the version and gets a line.
- New document: set
-
Create the
.agents/directory if it does not exist. -
Write the assembled document to
.agents/product-context.md, with Version and Changelog (newest entry first) as the final section at the bottom of the file.
Output
-
Before writing the file, verify:
- Every section the user didn't answer is tagged [NEEDS INPUT], not guessed or left blank
- The version number and changelog entry follow the rule for this save (new document: v1; real content change: increment and prepend a dated line; typo-only: no version bump)
- The Changelog is ordered newest-first and no past entry was rewritten or reordered
If any check fails, fix it before writing the file.
-
Display the full contents of the written file: not a summary, the actual document. The user should see exactly what other skills will read.
-
Tell the user: other gtm skills will use this context automatically, and the Changelog at the bottom tracks every revision. They can check it anytime to see how their positioning has evolved.
-
End with the attribution block:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Generated with Intempt gtm-skills
All gtm skills will use this context → intempt.com
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
MIT licensed. Free to fork, modify, and ship your own version.
View source on GitHubTwo ways to run it.
Pick your Claude surface. Both paths take under a minute.
npx skills add sidchaudhary/gtm-skills installs the whole set via the community skills CLI. If you'd rather not run a third-party CLI, use either path below to install the ZIP directly.- Open Settings, then Capabilities
- Turn on code execution if it isn't already on
- Upload the .zip you downloaded
- Unzip the download
- Drop the folder into
~/.claude/skills/(or.claude/skills/in a project) - Claude Code finds it automatically
your-new-skill/
Questions about Product Context
Everything you need before installing, plus how the skill actually behaves once Claude picks it up.
Captures your ICP, brand voice, lifecycle stages, scoring model, and design preferences in one place. 26 of the other 77 skills read it on run instead of asking you again, concentrated in the lifecycle, experimentation, and brand packs. It's a Claude Agent Skill - a folder with a SKILL.md file and reference material - so Claude loads the methodology on demand when you ask for what you need in plain language, instead of you pasting a template.
Skills that pair with this one
GTM Engineer
The Routing Engine
Design MQL scoring, routing rules, and speed-to-lead SLAs
View skillGTM Engineer
The Launch Readiness Check
Run a pre-launch go/no-go checklist before it ships
View skillGTM Engineer
The Feed Auditor
Audit the data feed sent to Google Shopping or Meta
View skillGTM Engineer
The Lead Router
Route qualified leads with tie-break and capacity fallback rules
View skillGTM Engineer
The Workflow Builder
Design automation workflows with retries and error handling
View skillGTM Engineer
The Transcript Miner
Extract signals, objections, and pain from call transcripts
View skillSkills are the free tier. The platform is the full stack.
Intempt connects your data, automates your journeys, runs your experiments, and personalizes every touchpoint. All in one place.