Weaver

Weaver is an open-source agent harness and orchestration framework maintained by sean35mm on GitHub. Built on top of the Bun runtime, it is designed to manage autonomous coding agent workflows with an emphasis on structured task lifecycle management and state persistence. In daily operations, Weaver coordinates multi-step development tasks by enforcing a scratchpad-first protocol. Agents write intermediate thoughts, execution plans, and context into dedicated scratchpads while persisting state and history across immutable SQLite/Bun database backends. Developers interact with Weaver through CLI scripts and inspect running tasks or agent memory via a bundled web dashboard. What distinguishes Weaver from simple prompt-loop harnesses is its hardened agent lifecycle architecture and built-in singleton dashboard for tracking agent activity. By treating task scratchpads and immutable database logs as primary coordination mechanisms, Weaver reduces context drift during extended refactoring and automated implementation runs. Weaver is published as an open-source tool distributed via GitHub under active pre-1.0 development (version 0.13.x).

Tags: agent harness, ai agent, cli, coding agent, developer tools, OpenSource

Visit Weaver

Key features

  • Scratchpads-First Protocol — Enforces structured agent scratchpad workflows to capture intermediate thoughts, state, and execution plans. - Bun-Native Runtime — Utilizes Bun for execution, script handling, and high-performance local database interactions. - Immutable Database Persistence — Stores execution runs and workflow state in immutable Bun-backed databases using URI configuration flags. - Singleton Web Dashboard — Provides a local browser interface (web/dashboard) to monitor agent scratchpads and execution lifecycles in real time. - Hardened Agent Lifecycle — Implements disciplined lifecycle controls to manage multi-step coding operations without context loss. - Claude and Multi-Agent Guidance — Integrates standardized agent guidance protocols (such as CLAUDE.md and AGENTS.md) directly into orchestration.

Use cases

Use Case: Persistent Multi-Step Refactoring Developers running long-running refactoring tasks can use Weaver to keep agents on track via structured scratchpads. The persistent state model ensures that plans, completed steps, and execution histories are not lost across context resets.

Use Case: Real-Time Agent Monitoring Teams running autonomous development scripts can launch the bundled dashboard to inspect scratchpad outputs and database logs. This provides direct observability into an agent's reasoning process without tailing raw log streams.

Use Case: Bun-Based Agent Scripting Engineers building lightweight TypeScript and Bun tooling can orchestrate agent tasks natively without the overhead of heavy Python-based harness dependencies.

Who it is for

Weaver is targeted at developers and systems engineers building autonomous AI workflows within the Bun and TypeScript ecosystem who need structured scratchpad tracking and local state persistence. It is less suited for non-technical users seeking a turnkey graphical IDE extension or a fully managed cloud service.

Install guide

Installation Clone the repository and install dependencies using Bun (assumed standard method for Bun-based repositories): bash git clone https://github.com/sean35mm/weaver.git cd weaver bun install

First Run Ensure Bun is installed on your system. Set up any necessary environment variables (such as provider API keys) and verify the setup: bash export ANTHROPIC_API_KEY="your-api-key" bun test

Example Session Start the agent workflow and launch the dashboard server: bash # Start the Weaver task orchestrator bun run src/index.ts # Launch the scratchpad dashboard bun run web/dashboard

Part of CLI Coding Agents

Categories: AI Coding Agents, Code Assistants, Developer Tools

Related CLI agents

  • SWE-agent — SWE-agent is an open-source autonomous software engineering agent harness built and maintained by researchers from Princeton University and Stanford University.…
  • AgentLint — AgentLint is an open-source repository linting, validation, and workflow enforcement harness developed by 0xmariowu. It is designed to enforce structural…
  • Context Bridge — Context Bridge is an open-source CLI agent harness and context orchestration utility developed by serdardb. It is built to facilitate…
  • Agentic Engineering Framework — Agentic Engineering Framework is an open-source orchestration harness and governance layer developed by Dimitri Geelen. It is designed to constrain,…
  • Hivelore — Hivelore is an open-source agent harness and contextual memory orchestrator created by Doucs91. It provides persistent project memory, decision tracking,…
  • Agent Runbook — Agent Runbook is an open-source contract-based multi-agent skill framework developed by KnoxOps. It acts as an orchestrator and compiler that…
  • AI Tools
  • Categories
  • Industries
  • CLI Coding Agents
  • MCP Servers
  • MCP Categories