The differentiator

A PM who reads
the repo
before the meeting.

Claude Code changed what a PM can own. I use it every day to do things that used to go to engineering or get skipped: trace GraphQL calls, search the repo, run survey analysis on a schedule, draft tickets that already sound like our team. The setup is deliberate: built over a year with context files, standing instructions, and automated workflows. It's the most distinctive part of how I work, and the thing I most enjoy teaching other PMs.


On this page
  1. I. How I use Claude day-to-day Setup · workflow · hard-won tips
  2. II. What I’ve written for others Training guide · portable prompt · Cowork

Part I

How I use Claude day-to-day.

Why it matters

Not a chatbot. A teammate.

Generic AI output is generic because there's no context. The trick — the only trick, really — is to invest in the context: a project folder Claude lives in, a CLAUDE.md it reads at the start of every conversation, a library of reference files capturing your team's voice and architecture, and MCP connectors so it can actually do things.

Once you've done that, Claude stops being "an AI chat" and starts being "a teammate that already knows the project, the team, and how you write." It drafts a ticket in your voice. It pulls release tickets via Jira and writes the release notes the way you would. It triages a Salesforce ↔ Hygraph integration question by reading the actual code. It synthesizes a folder of twelve interview notes and produces a themed write-up.

The investment in setup pays back inside a week. The longer you use it, the better it gets, because the context grows into a precise mirror of how you and your team actually work.

The setup

One folder. CLAUDE.md. Context files. MCPs.

A single project folder Claude lives in.

Everything related to Claude work goes here. Claude reads the file tree as part of its context, so a clean, predictable structure means it finds the right reference file every time without you having to point at it.

Claude projects/ ├── CLAUDE.md ← standing instructions ├── .env ← API keys (gitignored) ├── context/ │ ├── ticket-guidelines.md │ ├── formatting-strategy.md │ ├── product-architecture.md │ ├── team-directory.md │ ├── research-playbook.md │ └── exec-comms-style.md ├── ticket-templates/ ├── tickets/ ├── release-notes-drafts/ ├── trackers/ ├── comms/ ├── research/ │ ├── active/ │ └── archived/ ├── data/ └── repos/ ← local clones

CLAUDE.md as standing instructions.

The single most important file. It sits at the root of the project folder and Claude reads it at the start of every conversation. Six sections — overview, folder structure, communication preferences, code style, workflow, integrations note. Iterating on it is the difference between mediocre and excellent output.

## Preferences - Provide detailed reasoning and tradeoffs. - Present options with pros / cons before proceeding. - Ask clarifying questions if a request is ambiguous. - Match the tone of the destination doc, not your default. ## Workflow - Read existing context files before drafting. - Save outputs to the right subfolder, named clearly. - For Jira: output HTML, not Markdown. - For exec recaps: no em dashes, no bug fixes, business-outcome framing only.

Context files = your team's voice.

The context/ folder is where the magic happens. Markdown files that capture the institutional knowledge Claude needs to do high-quality work — what you'd brief a new hire on. Start with one. Add more whenever Claude misses something. After a few weeks you have a library.

  • ticket-guidelines.md

    Ticket structure, business-vs-implementation framing, word counts, section order.

  • formatting-strategy.md

    Jira Cloud → HTML. GitHub → Markdown. Confluence → either. Set once, followed forever.

  • product-architecture.md

    Integrations, key flows, terminology — Salesforce ↔ Compass ↔ Hygraph ↔ Auth0 ↔ Algolia.

  • exec-comms-style.md

    Shorter, no bug fixes, business outcomes only. Different voice than the internal release notes.

MCPs so Claude can actually do things.

Model Context Protocol = the connectors that let Claude talk to the systems your work lives in. Atlassian (Jira + Confluence), Bitbucket, the data warehouse, the CMS. Start with one — Atlassian alone unlocks 80% of the value for a PM.

  • Atlassian Read tickets, JQL search, fetch Confluence pages, post comments, create pages.
  • Bitbucket Read PRs, comment on PRs, open new PRs, list branches.
  • Hygraph CMS Query content via GraphQL. Trace UI logic to CMS-controlled config.
  • Data warehouse SQL queries against analytics. "How many users did X in 30 days?"
  • Figma Pull design context for tickets and handoff specs.
The workflow

A typical day, in prompts.

The setup above is the boring part. Here's what daily work looks like once it's in place — the actual prompts I run, and what Claude does in response.

My job got measurably better the day I stopped asking Claude what to do, and started telling it how I work.

Part II

What I’ve written for the next PM.

Teaching others

Three documents, built to travel.

Every time I figure out something worth knowing, I write it up — for my team, and for friends in other companies. These are the three that have traveled furthest.

Training guide · 18,700 words

Claude Code for PMs

A complete setup-and-workflow guide for non-engineering PMs. Five parts: overview, one-time setup, daily use, tips, and a five-week onboarding path. Built to hand to a new hire.

Read the guide
Portable prompt

Export your LLM context

I wrote this when I migrated from ChatGPT to Claude. It asks your old AI to summarize what it knows about you and produces three files: a context doc, a custom-instructions block, and a memory dump.

Read the prompt
Practical guide · ~80 sessions in

Cowork for the team

My actual Cowork setup. The automated meeting summarizer that runs every weekday at 5:36 PM. The weekly member-survey analysis pipeline. The connectors, the scheduled tasks, and how to build your own.

Read the guide
A few hard-won tips

Things I wish someone had told me at the start.

Want the full setup?

The full training guide is here, and I'm always happy to walk a PM through it on a call.

Read the full guide →

Or email me — meghan.race@gmail.com


Colophon

About this site, specifically.

Yes — the portfolio about how I use Claude was, in fact, built with Claude. It would have been a little awkward otherwise.

  1. Step 01 Claude · designer Voice, structure, art direction.
  2. Step 02 Claude Code · builder Every page, every line, every revision.
  3. Step 03 GitHub · source of truth Version control. Every iteration captured.
  4. Step 04 Vercel · host Push to main, deploy in seconds.
  5. Step 05 meghanrace.com · the domain Bought almost twenty years ago. Finally given a job.

A PM who can take a portfolio from concept to shipped in a long weekend — without an engineer in the loop — is the demonstration, not the credential. Welcome to the new working set.