Stop drowning in
meeting prep.
Works with the AI tools you already use
What changes
PM Pilot doesn't replace your judgment. It removes everything blocking it.
Meeting prep
45 min digging through Jira, Slack, and last week's notes
30 seconds. A brief lands in your terminal.
Weekly status
2 hours assembling updates, chasing people, guessing at progress
Auto-generated from real Jira data, formatted and ready to send.
PRD writing
Staring at a blank Confluence template for 40 minutes
Guided braindump-first process. Structure comes after thinking.
Market sizing
Days of spreadsheet hell, questionable data sources
Structured TAM / SAM / SOM in minutes. Cite your assumptions.
Start where you are
Never used a terminal? No problem. PM Pilot meets you at your current comfort level.
Works with any AI chat
Zero install
Copy any PM Pilot skill from GitHub and paste it into ChatGPT, Claude, Gemini, or any AI chat. No terminal, no setup. Works today, right now.
Better with Claude Desktop
5 min setup
Open Claude Desktop, create a project, and paste any PM Pilot skill into the project instructions. No terminal needed. Skills persist across every conversation.
Full power with Claude Code
Live Jira, Slack, Calendar
Install Claude Code CLI and connect your tools. PM Pilot reads live data: real tickets, real threads, real deadlines.
Five skills for your first week
Every skill is a markdown file you can read, edit, or fork. No black boxes.
Aggregates Jira tickets, Slack threads, and calendar context into a structured brief before any meeting.
Guides you through a braindump-first process before opening any template. No more blank-page panic.
Pulls real sprint data and generates a shareable status update for stakeholders. Formatted and ready.
Structures your TAM / SAM / SOM analysis with explicit assumptions. Works from a quick description.
Pulls meeting transcripts from Granola and builds a relationship map. Know who said what before your next conversation.
Example: /weekly-status in action
Three ways to get started
Pick the level that matches your comfort. You can always upgrade later.
Zero install
ChatGPT, Gemini, claude.ai
Go to claude.ai, ChatGPT, or Gemini. Paste the contents of any skill file into your conversation. That's it.
You won't get Jira or Slack integrations, but you'll instantly see the structured thinking PM Pilot brings to every task.
Try the meeting-prep skill fileClaude Desktop
5 minute setup
- Download the Claude Desktop app
- Open it and go to Settings → Projects
- Create a new project (e.g. "PM Pilot")
- Copy-paste any skill file contents into the project instructions
That's it. Claude Desktop remembers your skills between sessions. No terminal, no command line.
Full CLI with Claude Code
Maximum power
This is the full setup. Live Jira data, Slack threads, calendar context, meeting transcripts. Everything PM Pilot was designed for.
Follow the Claude Code setup guide, then run:
git clone https://github.com/mshadmanrahman/pm-pilot.git cd pm-pilot mkdir -p ~/.claude/skills ~/.claude/rules ~/.claude/agents ~/.claude/commands ~/.claude/memory cp -r skills/* ~/.claude/skills/ cp -r rules/* ~/.claude/rules/ cp -r agents/* ~/.claude/agents/ cp -r commands/* ~/.claude/commands/ cp memory/MEMORY-TEMPLATE.md ~/.claude/memory/MEMORY.md
Then run /configure-pm-pilot inside Claude Code to complete setup.
Don't have git? Download it here.
Connect your work tools
Level 3 only. PM Pilot gets dramatically more useful when it can see your actual work tools.
Jira / Confluence
meeting-prep, weekly-status, deep-context
Slack
meeting-prep, weekly-status, deep-context
Google Calendar
meeting-prep (knows your schedule)
GitHub
weekly-status, code context
Granola
people-sync (meeting transcripts)
No MCP servers connected? Skills like market-sizing, prd, prioritize, and critique work with zero external connections. You just type and go.
Memory that compounds
Most AI tools forget everything between sessions. PM Pilot builds a persistent knowledge base that grows with every conversation.
Memory directory
~/.claude/memory/
MEMORY.md - always loaded, under 200 lines
project_*.md - one file per project you own
feedback_*.md - corrections (so Claude doesn't repeat mistakes)
user_*.md - your preferences and working style
people/ - one file per key person
sarah-chen.md
marco-vidal.mdDay 1
Memory is empty. You explain your projects, your team, your preferences once.
Day 5
Memory has your projects, key people, and preferences. Much less explaining needed.
Day 15
Patterns and lessons are saved. Meeting prep pulls stakeholder context automatically.
Day 30
New sessions start with full context. You just say what to do. No backstory required.
Key insight: Every correction you make gets saved. You tell Claude something once. It does not repeat the same mistake.
Why this exists
PMs spend 60% of their time on status updates, meeting prep, and context-gathering. Not on product thinking. Not on strategy. Not on the work that actually moves the needle.
PM Pilot gives you that time back.
It started as one PM's personal setup. 14 years of product experience across startups and enterprise, distilled into reusable skills and workflows. The kind of system you'd build for yourself if you had the time.
Braindump before structure. Never open a template before the thinking is externalized.
Memory over transcripts. A 200-line memory file beats a 50,000-token session replay.
Skills load on demand. 25 skills, zero startup cost.
Rules are compressed. Around 950 tokens total, not 9,500.
Every session compounds. Corrections become rules. Meetings become stakeholder intelligence.
Full skill reference
25 skills, 5 agents, 4 slash commands. All open source, all editable.
PM Core (12 skills)
Briefing from Jira, Slack, and Calendar. Includes political context and open questions.
Reads Granola transcript, updates stakeholder files with new context.
Accomplishment report pulled from connected systems. Formatted and ready to send.
Cross-channel research across Jira, Slack, Confluence, and memory.
TAM, SAM, SOM with explicit data sources and assumptions.
Enterprise knowledge assistant. Searches org context and memory.
Systematic QA walkthrough with structured bug reports.
Searches 269+ PM podcast episodes for relevant insights.
Braindump first, then structure. No more blank-page paralysis.
RICE, ICE, WSJF, MoSCoW, Kano. Pick your framework or let it recommend.
Extracts themes and actionable recommendations from user research.
Pressure-tests documents for logic gaps, missing assumptions, and weak arguments.
Productivity (6 skills)
Loads workspace context at session start
Creates portable handoff documents
Compresses strategy into actionable briefs
Wave-based parallel agent execution
Reads and summarizes agent manifests
Background observation and logging
Dev (4 skills)
Test-driven development enforcement
Build, test, lint verification
Research before implementation
Security audit for code changes
Content (3 skills)
Competitive and market analysis
Personal voice and tone enforcement
Substack-optimized content creation
Slash commands
Agents
Ready?
Free. Open source. No account needed.