project-butler

English | 中文

Make AI coding agents remember your project between sessions.

project-butler helps Claude Code, Cursor, Codex, and similar AI coding assistants behave like long-term project teammates instead of starting from scratch every session.

For normal use, you only need four actions:

/project-butler   Set up project memory
end session       Save progress and next steps
continue          Resume next time
status            Check where the project stands

For projects that need stronger product, architecture, roadmap, research, or eval alignment, project-butler can also create a Project Profile during setup and offer profile-aware Normal Close / Full Close behavior.

Quick Start

Install as a Claude Code skill:

git clone https://github.com/JamesShi96/project-butler.git ~/.claude/skills/project-butler

Open any project and set up project memory:

/project-butler

Work normally. At the end of a work session:

end session

Next time, resume without re-explaining the project:

continue

That is enough for daily use. For Cursor, Codex, and other assistants, see Tool Compatibility.

Why It Exists

AI coding assistants are powerful in one session and forgetful across sessions. If any of these sound familiar, project-butler is for you:

• "I had to re-explain the architecture again." Each new session starts with missing context.
• "What did we decide about naming conventions last week?" Decisions disappear into chat history.
• "The README and TODOs keep drifting from reality." Project state stops matching the files.
• "The AI keeps violating rules I already explained." Rules live in your head instead of in project memory.
• "I switch between Claude Code, Cursor, and Codex." Different tools need one shared source of truth.

project-butler turns a project folder into that source of truth, so the next AI session can pick up where the last one stopped.

Main Commands

All triggers are natural language. Use slash commands only for first-time setup.

Advanced Commands

Session recovery (continue / continue full context) is routed through project-butler internally. There is no separate /continue command to install.

What It Maintains

Run /project-butler once. It maintains these plain Markdown files in your project:

project-root/
├── CLAUDE.md                   <- Project rules / constitution
├── PROJECT.md                  <- Current project wiki
├── STRUCTURE.md                <- File organization rules
├── UPDATE_LOG.md               <- Milestone-level changelog
├── DOCS.md                     <- Document index and metadata
├── session-handoff.md          <- Cross-session handoff
├── TODO.md                     <- Execution checklist
├── docs/                       <- Archived project documents
├── log/                        <- Session logs
└── .claude/
    ├── candidates.md           <- Candidate rules for review
    ├── project-profile.json    <- Optional project profile config
    ├── profile-pending.json    <- Optional profile pending/debt queue
    └── .file-snapshot.json     <- File organization snapshot

The core files are plain Markdown, so other tools can read them even when they do not run the skill natively.

What that means in practice:

• Keeps the current project state readable.
• Keeps next steps clear between sessions.
• Keeps project documents indexed and findable.
• Records milestone changes so the project has a clear history.
• Keeps new files from drifting into random folders.
• Preserves long-term rules only after user review.

When Project Profile System is enabled, it also keeps a small machine-readable profile so the assistant can understand which long-lived docs matter, which sections are protected, and which profile updates have been deferred.

Tool Support

See docs/compatibility.md for details and caveats.

How It Works

Internals: The Memory Stack

project-butler uses a 7-component memory stack internally, organized by stability:

Stable rules
┌─────────────────────────────────────┐
│  CLAUDE.md / project rules          │  <- Human-reviewed principles
│  ↑ candidates collected by AI       │
└─────────────────────────────────────┘
            ↑ distilled from work
Current state
┌─────────────────────────────────────┐
│  PROJECT.md                         │  <- What the project is now
│  STRUCTURE.md                       │  <- Where files belong
│  UPDATE_LOG.md                      │  <- Milestone-level changes
│  DOCS.md                            │  <- Document index
│  .claude/project-profile.json       │  <- Optional profile config
│  .claude/profile-pending.json       │  <- Optional profile debt queue
└─────────────────────────────────────┘
            ↑ summarized from facts
Raw facts
┌──────────────────────┐ ┌───────────────────────┐
│  log/                │ │  TODO.md              │
│  What happened       │ │  What needs doing     │
└──────────────────────┘ └───────────────────────┘
            ↓
session-handoff.md       <- Where the next session should resume

Bottom feeds top. Top constrains bottom.

• Session logs capture what happened.
• Handoff tells the next assistant where to resume.
• Project wiki summarizes the current state.
• TODOs keep execution visible.
• Rules / constitution preserve decisions that should keep guiding the project.
• Update log records significant changes at milestone level.
• Structure rules keep files from drifting into chaos.
• Document index keeps project documents organized under docs/.
• Project profile optionally tracks project shape, document tiers, document policies, pending profile updates, and review queue items.

Project Profiles

Project Profile System is optional profile-aware behavior for projects that need stronger long-term alignment.

During setup, project-butler can ask what you are trying to do in natural language, infer the project shape, ask a few targeted follow-up questions, and propose Required / Recommended / Optional reference docs. It does not force you to pick a fixed project type.

During close, profile-aware projects can use:

Full Close confirms boundaries, not every small edit. It still requires explicit confirmation before changing protected sections, document policies, stable baselines, or whole-document rewrites.

Language Support

project-butler supports three language modes:

You choose the mode during setup, and can later say change language.

Version Naming

During setup, project-butler asks which version style the project should use:

end session uses this style when a significant update deserves an UPDATE_LOG.md entry.

Upgrade Mode

If a project already has some management files, project-butler creates only the missing ones. It does not replace existing files; when a system section needs an update, it asks before making a small targeted patch. It also detects legacy .claude/memory/ layouts and suggests migration.

Profile System is not forced during ordinary upgrade. If a project already has profile files, project-butler preserves them. If the user asks for profile setup, full close, or foundation repair, project-butler can infer a profile from existing project docs and ask for confirmation before writing profile files.

Examples

See docs/examples.md for a complete session flow:

1. initialize a project,
2. work normally,
3. end the session,
4. resume the next day,
5. review accumulated rules.

Requirements

• Claude Code CLI for native skill execution
• jq for continue / continue full context session recovery
• Optional: Cursor for generated project rules
• Optional: Codex or other AI coding assistants that can read project Markdown files

Update Log

v1.6.0 (2026-06-10) - Project Profile System Runtime Wiring

• Add references/project-profile-system.md and route profile setup, Normal Close, Full Close, Foundation Repair, and profile-aware status through the main skill.
• Teach generated project rules, continue, full context recovery, and upgrade mode to read and preserve optional profile files.
• Document Project Profile System as optional profile-aware behavior, not part of the base 7-component memory stack.

v1.5.1 (2026-06-03) - Product Noise Reduction

• Reframe the README around four primary actions: /project-butler, end session, continue, and status.
• Move internals behind "How it works" and make examples result-focused.
• Update generated CLAUDE/Cursor rules to separate daily workflow from advanced commands.

v1.5.0 (2026-06-03) - Versioned Update Log System

• Add version style selection during setup: Semantic, Codename, Patch, and Date.
• Teach end-session update logging to calculate the next version from the selected style.
• Sync README, examples, compatibility docs, generated rules, continue recovery, document archiving, upgrade mode, and trigger routing with the current memory stack.

v1.4.1 (2026-06-02) - Cross-Reference + Flow Consistency

• Fix DOCS.md, file snapshot, language switching, continue recovery, and template consistency gaps found through six review rounds.
• Align document archiving, file reorganization, and Cursor rule templates.

v1.3.0 (2026-06-01) - Four-Phase File Reorganization

• Replace template-based file organization with Discover, Ask or Plan, Plan, and Execute phases.
• Add confidence routing, safe move planning, and a never-delete cleanup policy.

v1.2.1 (2026-05-09) - Skill Loader Compatibility

• Move continue reference workflows under references/ instead of nested SKILL.md files.
• Shorten skill metadata to satisfy loader limits.
• Keep session recovery routed through the main project-butler skill.

Full update log: UPDATE_LOG.md | Releases: GitHub Releases

Star History

License

MIT