Skip to main content

Collaborative file locking runtime

Project description

collab-runtime — Collaborative File Locking

Prevent merge conflicts by automatically locking files when a developer starts editing them, using Supabase Realtime as the backend.

Python License: MIT PyPI


What It Does

collab-runtime provides a CLI and background daemon that coordinate file-level locking across a development team in real time. When a developer opens a file, a lock is acquired in Supabase. Other developers are immediately notified — via desktop notifications, VS Code warnings, or the web dashboard — that the file is in use.

Key features:

  • Atomic lock acquisition — Supabase RPC prevents two developers locking the same file simultaneously
  • 📡 Real-time broadcast — Supabase Realtime pushes lock events to all connected clients instantly
  • 🖥️ Web dashboard — Live lock status view, force-release controls
  • 🔔 VS Code extension — Lock warnings, status bar, output channel
  • 🪝 Git hooks — Pre-commit and pre-push hooks block commits of locked files
  • 📋 Audit trail — Full lock history with configurable retention

Prerequisites

Requirement Version / Notes
Python 3.10 or higher
Supabase account supabase.com — free tier works
Node.js Only required for the VS Code extension

Installation

pip install collab-runtime

For a minimum-version install (ensures you get the latest compatible release):

pip install "collab-runtime>=0.3.2"

Quick Start

1 — Create the Database Schema

In your Supabase project, open SQL Editor and run the contents of supabase/schema.sql.

This creates the file_locks table, file_locks_history audit table, the atomic acquire_lock() RPC, Row Level Security policies, and Realtime publication.

2 — Configure Environment Variables

Create a .env file in your project root:

SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=your_anon_key_here
SUPABASE_SERVICE_ROLE_KEY=your_service_role_key   # Required for force-release
DEVELOPER_ID=your_name                             # Optional, defaults to git user.name
LOCK_STRICT=0                                      # 1 = block on lock errors, 0 = warn only
COLLAB_AGENT_ID=agent-my-task                      # Optional: unique id per AI agent session
COLLAB_AGENT_LABEL=refactor-auth                   # Optional display label
COLLAB_AGENT_MODE=1                                # Auto-generate/persist agent id when unset

Keep SUPABASE_SERVICE_ROLE_KEY private — never commit it to version control.

Multi-agent usage (same GitHub user, multiple AI agents)

When one developer runs several AI agents in the same repo, give each agent its own identity so locks do not collide (effective owner is developer_id + agent_id):

# Agent A
export COLLAB_AGENT_ID=agent-refactor-auth
collab whoami
collab acquire src/auth.py --reason "Refactor auth"

# Agent B (different id) — conflicts with A on the same file
export COLLAB_AGENT_ID=agent-fix-tests
collab acquire src/auth.py

CLI flags --agent-id and --agent-label override env vars. Use collab active --mine to list only locks held by the current human + agent pair.

Existing Supabase projects: apply the agent_id / agent_label columns and updated acquire_lock function from supabase/schema.sql in the SQL Editor (fresh installs already include them).

3 — Verify Connection

collab active

If connected, this lists all currently active locks (empty on a fresh setup).


CLI Reference

# Show resolved developer and agent identity
collab whoami

# Show all active locks across the team
collab active
collab active --mine

# Lock a file before editing
collab acquire path/to/file.py --reason "Implementing feature X"

# Release a lock when done
collab release path/to/file.py

# Check lock status for a specific file
collab status path/to/file.py

# Release all locks held by you
collab release-all

# Force release (requires SUPABASE_SERVICE_ROLE_KEY)
collab force-release path/to/file.py
collab force-release-all

# Batch operations
collab acquire-batch path/to/a.py path/to/b.py --reason "Refactoring"
collab release-batch path/to/a.py path/to/b.py

# Reconcile local and remote lock state
collab reconcile

# Lock history and retention
collab history
collab history path/to/file.py --limit 50
collab history-prune --days 30

# Background watcher daemon
collab daemon-start
collab daemon-start --interval 10 --timeout 480
collab daemon-status
collab daemon-stop

# Foreground watcher (diagnostics)
collab watch --interval 5 --timeout 0

# Web dashboard (opens in browser)
collab dashboard

# Cleanup orphaned watcher processes
collab cleanup

VS Code Extension

The optional VS Code extension provides lock-on-open warnings, a status bar indicator, and one-click dashboard access.

Install from source (extension is bundled in the GitHub repository):

  1. Press F1Developer: Install Extension from Location...
  2. Select the editors/vscode/collab-locks/ directory
  3. Reload VS Code

Once installed, the extension automatically starts and stops the background daemon with your VS Code window.


How It Works

Developer opens file
      │
      ▼
collab acquire (CLI or extension)
      │
      ▼
Supabase RPC: acquire_lock()   ◄─── atomic, unique constraint
      │                                prevents double-locking
      ▼
Supabase Realtime broadcast
      │
      ├─► VS Code extension  →  lock warning popup
      ├─► Web dashboard      →  live status update
      └─► Desktop notification (via plyer)

Lock release follows the same path in reverse and writes an entry to file_locks_history for audit.


Security

  • Lock correctness is enforced at the database level via atomic RPC — no client-side race conditions.
  • Force-release requires SUPABASE_SERVICE_ROLE_KEY; regular releases are scoped to the owning developer.
  • Row Level Security (RLS) is configured on all tables via supabase/schema.sql.
  • Never commit secrets; use .env for local configuration only.

Links


License

MIT — see LICENSE for details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

collab_runtime-0.5.0.tar.gz (109.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

collab_runtime-0.5.0-py3-none-any.whl (107.8 kB view details)

Uploaded Python 3

File details

Details for the file collab_runtime-0.5.0.tar.gz.

File metadata

  • Download URL: collab_runtime-0.5.0.tar.gz
  • Upload date:
  • Size: 109.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for collab_runtime-0.5.0.tar.gz
Algorithm Hash digest
SHA256 8be8bbb4656e56eab363ef2268899e64c878f7db803b87bbb0ebdfb34b6bd7ae
MD5 56d934d4f6eee4f343a57bc71000e881
BLAKE2b-256 9aae92fc2c0ef85949f1dfeab9107affd7e994675fec8178e7cbc9840c28fd76

See more details on using hashes here.

Provenance

The following attestation bundles were made for collab_runtime-0.5.0.tar.gz:

Publisher: publish.yml on KirilMT/collab

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file collab_runtime-0.5.0-py3-none-any.whl.

File metadata

  • Download URL: collab_runtime-0.5.0-py3-none-any.whl
  • Upload date:
  • Size: 107.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for collab_runtime-0.5.0-py3-none-any.whl
Algorithm Hash digest
SHA256 a3a9058957d22e1c0f0e46d000b34ae3716911856e7a7cf5c4d19cd1072d0078
MD5 3b57610740625f4fb693f99f597a2420
BLAKE2b-256 1ab08a946b458fa27c836ae48f79d590fee2e56d560dfc8ad58104a2e8384647

See more details on using hashes here.

Provenance

The following attestation bundles were made for collab_runtime-0.5.0-py3-none-any.whl:

Publisher: publish.yml on KirilMT/collab

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page