Weft

Weft is an open source protocol and coordinator that manages concurrent coding agents during edits.

Weft coordinates coding agents while they edit code repositories instead of waiting for pull requests. Agents submit tool-call edits to a per-repository sequencer that validates changes against accepted history and returns diagnostics. Humans view the event stream and can approve, pause, message, undo, or inspect evidence in real time. The architecture uses Cloudflare Workers, Durable Objects, D1, R2, and sandboxes to isolate git and test operations. It includes adapters, an analyzer, workflows, and a web interface under the MIT license.

Weft at a glance

Best for: Software engineer
Runs on: Web, self-hosted, open source
Pricing: Free
Released: October 2026
Category: AI coding tool
Website: github.com/celador/weft

Key features

Submits tool-call edits to a per-repository sequencer
Validates edits against accepted changes
Returns actionable diagnostics to agents
Displays a live event stream for human monitoring
Isolates git and test operations in sandboxes
Indexes tasks and evidence using D1
Stores run artifacts and logs in R2

Use it for

Coordinating multiple coding agents editing simultaneously
Inspecting real-time event logs of agent runs
Reviewing and approving agent code modifications
Running protocol conformance tests locally

Companions

Tools that articles about Weft recommend using with it, each with its job.

Context7 — Provides up-to-date documentation for AI agents and code editors.
Ollama — Run local models at zero marginal cost, with code kept on your machine.
OpenRouter — Access over 500 models and 80 providers through one unified API.
Daytona — Runs sessions in isolated sandboxes synced to git branches.
Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps
Oh My OpenCode Slim — Lightweight multi-agent orchestration and background tasks.

Alternatives to Weft

jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
OpenCode — Works with 75+ model providers, local models, and existing subscriptions
GitAgent — Manage AI agents with familiar git workflows like branching, diffing, and reverting.
Warp — AI terminal plus cloud agents with built-in evals and benchmarks
CodeGraph — It eliminates file crawling, reducing AI agent tool calls by 88% and cutting costs.
Dirac — Executes complex coding tasks efficiently with hash-anchored edits and low token overhead.
Invariant — Persistent isolated Linux environments and parallel coding agent experiments.
Woml — Combine readable HTML-inspired markup with JavaScript logic and managed durable execution.
Crit — It keeps human feedback localized and precise for AI agent code iterations.
OpenAgentsControl — Generates code matching your patterns and requires approval before execution.
Sphinx — Requires understanding of agent-generated code before shipping.
Mindwalk — Visualize how coding agents explore and edit files on a 3D repository map.

All Weft alternatives