pm-oss-launch

Use when asked to plan a short demo video for a repo, script a 20-second clip, storyboard a GIF for a README or launch post, or make a project look good on X or LinkedIn. Produces the script for the 20-second video that sells a repo: a shot list with a timestamp per shot, on-screen text, exactly what to record, the hook in the first two seconds, and post copy for X and LinkedIn. For a live product demo to an audience, use demo-script; for cutting clips from a long recording, use clip-factory.

Use when asked to create demo data, seed a no-signup demo, generate fake but realistic data for screenshots, or fill a sandbox without using real customer data. Produces realistic, internally consistent fake data so a project can ship a demo nobody has to sign up for: a data schema summary, a seeded generation script, seed files, and a checklist confirming no real personal data is included.

Use when asked to create good first issues, prepare a repo for new contributors, turn a roadmap into starter tasks, or write issues that newcomers can actually finish. Produces 8 to 12 good first issues from a codebase and roadmap, each with context, acceptance criteria, files to touch and an effort estimate, plus the labels and a short pinned welcome note. For the CONTRIBUTING guide itself, use contributor-guide.

Use when asked to plan releases for a side project or small open-source repo, turn a backlog into a release schedule, ship smaller and more often, or write a ROADMAP.md. Turns a backlog into small themed two-week releases and produces ROADMAP.md, a theme and scope for each release with what is deliberately left out, and a changelog template. For a stakeholder roadmap narrative inside a company, use roadmap-narrative.

Use when asked to build an importer from a competitor's export, migrate users' data from another tool, map a CSV or JSON export onto my data model, or make switching to my project easy. Given a sample export (CSV or JSON), produces a field mapping table, an importer code skeleton, a rule that unmatched records are kept rather than dropped, and test fixtures including the awkward cases. For a database migration inside one system, use database-migration-plan.

Use when asked to check a project's licences, audit dependencies before open-sourcing or selling, write a NOTICE file, find licence conflicts, or check whether bundled fonts, images or datasets can be shipped. Produces an inventory of code, data and media dependencies with their licences, flags for conflicts and unclear ownership, and a draft NOTICE.md, with a clear statement that the output is not legal advice.

Use when asked to rewrite a README so strangers understand why they should care, turn a feature list into benefits, make a side project's README convincing, or explain what users get rather than what the code does. Rewrites an existing README so every feature is stated as a user benefit in plain sentences, and produces a one-line pitch, a Why section, a benefit-led feature list, a quick start, a Your data section and a contribution pointer. For a README written from scratch with badges, usage and install sections, use readme-writer.

Use when asked to expose a folder of notes or docs to Claude or another AI client, build an MCP server over Markdown or JSON files, let an agent read my knowledge base safely, or wrap documentation as MCP tools. Produces a read-only MCP server over a folder of Markdown or JSON files: server code using stdio transport, tool definitions, a client configuration snippet, and a safety section confirming no write, delete or network tools are exposed. To design an MCP server for a whole product, use mcp-server-spec.

Use when asked to make a project self-hostable, package an app for Docker, write a docker-compose setup, document self-hosting, or let users run my app on their own server. Produces a one-command self-host for a project: docker-compose.yml, a .env.example with a table explaining every variable, SELF_HOSTING.md covering HTTPS, backups and upgrades, and a back-up-this-folder section that tells users exactly what holds all their data.