Build your GTM context brain this week
Episode 2, made runnable. Context is the product, but bad context in is bad output out. Define your ICP, positioning, and voice first, then build the repo any agent can run on. Already done run-your-gtm-on-github? Skip its 4 repo-mechanics steps and start at step 1 here.
Clone the Sales Operator starter kit.
Do not start from an empty folder. Clone the Sales Operator starter kit: a five-layer context pack with the CLAUDE.md brain, every context template, the loop and context skills, workflows, and a fully worked example. Substance-first, runnable on clone.
New to GitHub? Run the github-for-gtm skill first.
Build your GTM context brain this week
Episode 2, made runnable.
Context is the product, but bad context in is bad output out. Define your ICP, positioning, and voice first, then build the repo any agent can run on. Already done run-your-gtm-on-github? Skip its 4 repo-mechanics steps and start at step 1 here.
Details
Details
Details
Details
Details
Details
Details
Details
The governance file
The GTM repo itself, CLAUDE.md and all, with the QA gates that keep it from shipping broken in production.
Your ICP
Positioning
Voice defined
Then structured into a CLAUDE.md brain
Most teams rush to build the repo before they have defined their ICP, positioning, or voice. Bad context in is bad output out. The tool was never the problem, and neither was the folder structure. The missing, undefined context was.
The workflow that solves it, one step at a time.
Each step is the plain question you are already asking. The skills answer them in order, one handing its receipt to the next.
Names your tiers with explicit criteria, the fit signals, the anti-ICP you exclude on purpose, and a qualification framework, plus an evolution log. A category is not a definition.
Turns what you do, who it's for, and why you win into one positioning statement and a full message house.
Sets your voice pillars, tone by context, a do and do-not word list, and sample lines the model can pattern-match, so the whole repo writes in your voice.
The operator version of git: the mental model (repo, commit, branch, push, pull), a GUI-first workflow so you never fear the terminal, how to organize the repo, and the hard rules on what to never commit. The prerequisite nobody teaches GTM.
A single pack that tells a new session or teammate what the project is, where things live, what's decided, and the next move.
A plain-text, searchable base of facts, playbooks, stakeholders, and how-tos, organized so the answer is one search away.
A proposed naming convention, folder structure, index, and cleanup plan, shown as a plan first so you see every move.
A running log where each entry captures the decision, the context, the options, the why, and when to revisit.
Your ICP, positioning, and voice defined, then structured into a CLAUDE.md brain and a context pack any teammate or agent runs on, kept current.
- All 8 chained skills resolve to real, scored skills.
- Has a problem, an outcome, 2+ steps, and proof from a real build.
- Ships an example chat showing the play run in Claude.
- Reads in plain operator voice, grounded in a real build.
We did not just check the structure. A judge read this play against the full text of every skill it chains (composition), then ran it on a real scenario end to end (simulated). It passed both. Test scenario: A solo GTM lead at a 20-person startup has never used GitHub and has no written ICP, positioning, or voice guide. She wants a real Claude Code context repo by Friday.
This play runs across the org, not inside sales.
Every function below owns a piece of the play. Skip one and it stalls where most GTM plays stall: in the handoff.
Without it: Every agent output ships in a different voice.
Without it: The repo encodes an ICP the market already left.
Without it: Agents confidently pitch features that do not exist.
Without it: The brain and the CRM tell different stories.
Watch the skills answer in order.
Describe your situation in plain language. The right skill fires on its own and hands its receipt to the next.
Before you run it, I ran it for real.
This workflow is genericized from a real build. Here is the number it moved, and what I actually did.
WHAT I DID ›
When you automate, your failures go silent. Write them down as enforced gates, not documentation. A governance file that reads your own failure history is how a pipeline stops repeating its worst week.
Dig into the full build →Context rots unless someone owns the pruning
The skills hand the team an ICP, positioning, and voice structured into a repo any agent can run on; the value comes from a teammate getting a trustworthy answer without asking anyone. Here is how a GTM leader keeps the repo current after the first week's energy fades.
Once a week, one owner walks the repo asking a single question of each file: would an agent acting on this today be wrong? Anything stale gets fixed or dated, and every material call lands in the decision-log the day it is made.
- 01Teach what good context looks like
Show the team the difference between a file an agent can act on and a paragraph of vibes: tiers with criteria, positioning said the same way every time, voice with examples. The bar is actionable, not thorough.
- 02Set the same-day decision SLA
A call that changes the ICP, pricing, or positioning goes into the decision-log the day it happens. A decision that lives only in a meeting is a future re-litigation.
- 03Rehearse the cold start
Have a teammate or a fresh agent run a real task from CLAUDE.md alone, no verbal context. Where they stumble is where the repo is wrong, whatever the file says.
- 04Sweep for rot on a cadence
Run the stale-context sweep weekly and a full pass with workspace-organizer quarterly. The repo is a system of record; treat it like one.
One plugin. 8 skills, bundled.
The whole playbook installs as a single plugin in Claude Code, no copy-pasting 8 times. Or grab any single step above. It runs on what you paste; connect your stack to go live.