Pad

Pad is a local-first project management tool built for individual developers and the AI agents that work alongside them. It ships as a single Go binary that runs on your machine, with a SvelteKit web UI on localhost:7777, a pad CLI, and a /pad skill that integrates with Claude Code, Cursor, Windsurf, Codex, OpenCode, GitHub Copilot, Amazon Q, and JetBrains Junie. All data lives in a SQLite file on your laptop; Pad Cloud is an optional sync and sharing layer if you want multi-device or team workspaces.

Pad organises work into typed collections. Built-in collections cover tasks, ideas, plans, docs, conventions, and playbooks, and you can define your own collections with custom fields (select, text, date, number, URL, relation, checkbox). The web UI offers board, list, and table views with drag-and-drop, keyboard navigation, a Tiptap-based rich text editor with markdown, wiki-style [[links]], and real-time updates from the CLI via SSE.

The agent integration is what sets Pad apart. Conventions are trigger-based rules ("run tests before marking a task done"), and playbooks are multi-step workflows the agent can follow on demand — for example /pad ship PLAN-42 or /pad release 0.5.0. Running /pad onboard in an agent session inspects your codebase, detects build and test tooling, and adapts collections, conventions, roles, and playbooks to match the project. Every agent action is attributed in the activity feed.

Compared to Linear, Jira, and Notion, Pad is fully offline-capable, requires no account, and is priced only by optional cloud usage. It is best suited for solo developers and small teams who want their project data local and want their AI coding tools to act on the board rather than just read from it.