PM Brain skill
A Claude Code skill collects product context in Markdown and links decisions to interviews, documents and hypotheses
- 954 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Initialization writes files, hook settings and local Git history
- The brain retains work documents and stakeholder information read by the model
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/pm-brainDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Initialization writes files, hook settings and local Git history
- The brain retains work documents and stakeholder information read by the model
This is third-party code. Review the repository files before installing.
What it does
The PM Brain skill helps Claude Code create a knowledge base for one product in a chosen directory. Claude Code first gathers company context and current priorities, then copies a file scaffold and fills it from answers or existing material. The agent keeps each original document separately, writes a synthesis and updates related knowledge, hypotheses and decisions. Provenance markers distinguish documentary evidence from verbal comments or assumptions. Project instructions then support ingestion, meeting preparation and weekly review. The system reads and edits ordinary files without a vector database or a separate model-memory service.
Who it is for. Product managers retaining decision evidence and revisiting it after new interviews and observations.
Good fit when
- Consolidate one product's context from interviews, documents and decisions
- Prepare for a stakeholder meeting using their unresolved questions
- Propagate new information that contradicts an earlier hypothesis
- Review stale conclusions and evidence gaps during a weekly sweep
Not a fit when
- You need automatic synchronization of every work service without separate connections
- You cannot retain interview copies or stakeholder information in project files
- You want setup to overwrite an active repository without reviewing changes
Example request
Set up PM Brain in a new directory for my product. Ask about current priorities and show the structure first. Preserve interview sources separately and mark unsupported conclusions as hypotheses.Limitations
This research preview’s synthetic checks do not establish behavior in every organization. Setup copies hidden Claude Code settings and can create a local Git commit; inspect the destination first. The validate_brain_file.py hook needs Python 3 and checks structure and references rather than factual truth. Documents and stakeholder information reach the selected Claude model; Markdown storage does not make the workflow fully local.
How to disable. Remove the global pm-brain directory from Claude Code's skills location. Separately disable the created project's hook in `.claude/settings.json` if no longer needed. Knowledge files, source documents and Git history remain in the working directory.
FAQ
How does installing the skill differ from creating a brain?
Install the skill globally once. `/pm-brain` then initializes a brain in the current work directory. Another product can use another folder.
How do I add an interview?
Use `/ingest` with a file or pasted text. Instructions keep untouched input in `source/`, synthesis in `ingestion/` and update related records.
How does meeting preparation work?
`/prep <stakeholder>` produces a brief with open requests, the last unresolved concern and suggested questions. Match the name to a stakeholder record.
Does the weekly review run as a background schedule?
No. Invoke `/review` in Claude Code for a review and proposed changes. Choose the cadence during setup; no separate background service is installed.
How are decisions recorded?
Entries under `decisions/` retain evidence and reopening conditions. A new signal can trigger an update. Having a link alone does not establish that it supports the conclusion.
What license does PM Brain use?
MIT.
Discussion
No questions or reviews yet.
Related
SkillsSecond Brain Skills
Claude Code Second Brain Skills
A set of Claude Code skills for knowledge work: brand and voice, an MCP client, PPTX slides, Remotion video, SOPs and a skill creator
PM Skills
PM Skills Marketplace
A plugin marketplace for product managers: discovery, strategy, PRDs, launch, growth and audits of AI-built code
A set of 13 SKILL.md skills: research, a product doc, decks, articles and audits for any compatible agent
Claude Code Toolkit (robertguss)
Claude Code Toolkit
A set of Claude Code skills and hooks: brainstorming, worker-led project leadership, a wiki, and a book pipeline.
A skill that keeps project memory in plain Markdown files: session logs, a wiki, rules, a to-do list and a note on where to resume
A product management skill library for agents: product phases, sprints, templates and sample outputs
README in short
A filled example brain and week-long walkthrough illustrate use. The same author's PM Skills adds specific product-management methods through separately installed plugins.
README badge
Original SKILL.md
--- name: pm-brain description: Initialize a PM Brain — a markdown-native second brain for a product operator (PM, product lead, founder, or anyone accountable for one product or initiative) doing judgment-heavy work with scattered inputs. Detects greenfield vs. migration mode, runs a focused interview, copies the deterministic scaffold into the working directory, populates placeholders from interview answers, runs a self-test, and commits. Use when invoked via `/pm-brain` or when the user asks to set up a PM Brain. --- # PM Brain — Skill This skill scaffolds and initializes a PM Brain in the current working directory. The scaffold is deterministic (static files copied as-is from `scaffold/`). The reasoning is adaptive (loaded from `prompts/` per phase). ## Architectural split | Layer | Where it lives | Why | | --- | --- | --- | | **Static structure** — schemas, CLAUDE.md, INDEX.md, folder tree, file templates | `scaffold/` | Deterministic. Same every time. No generation needed. | | **Adaptive reasoning** — mode detection, migration, interview, post-scaffold self-test | `prompts/` | Probabilistic. Depends on what's in the directory and what the PM says. | | **Orchestration** — when to do what | This file | Glue. | Behavior evolves independently from structure. Schemas can change without touching reasoning. Reasoning can improve without rewriting schemas. ## When to invoke - Operator runs `/pm-brain` (or pastes a setup request like "set up a PM Brain here"). - Operator asks to add a PM Brain to an existing directory of PM artifacts. Do **not** invoke this skill for routine PM Brain operations after init (ingestion, prep, review). Those are handled by the seeded `CLAUDE.md` operating manual in the target repo. ## Workflow ### 1. Detect mode Load `prompts/mode-detection.md`. Inspect the current working directory. Decide: **greenfield** (empty), **migration** (PM artifacts present), or **active-repo** (working repo — pause and ask). Announce the detected mode to the operator in one line. For active-repo mode, do not proceed without confirmation. ### 2. If migration mode Load `prompts/migration.md`. **Copy** (do not move) pre-existing PM artifacts into a `source/` folder. Bulk-ingest with epistemic caution. Record cross-document conflicts for the post-scaffold contradictions block. ### 3. Run the interview