Codebuff
Also known as: Codebuff AI, Freebuff, Manicode, CodebuffAI, Codebuff, Inc.
Open source terminal coding agent whose work is split across named file-picker, planner, editor and reviewer agents, with per-agent model choice on OpenRouter, a TypeScript SDK for embedding, and a public Agent Store of reusable agents.
Codebuff is an open source, terminal native AI coding agent whose distinguishing idea is a multi agent architecture.
Rather than asking one model to do everything, it coordinates specialized agents that each own a step: a file picker agent scans the codebase to understand its architecture and find relevant files, a planner agent decides which files change and in what order, an editor agent writes the code, and a reviewer agent validates the result.
The pitch is that splitting the work this way gives better context, more accurate edits, and fewer errors than single model tools, especially on architectural, multi file changes. The project is published under Apache-2.0 with a public repository carrying roughly 8,300 stars and over 8,000 commits.
Codebuff, Inc. was founded by James Grugett and Brandon Chen in San Francisco and backed by Y Combinator in its Fall 2024 batch. The product launched originally as Manicode in 2024 and was renamed to Codebuff after a trademark dispute.
Codebuff is built for wholesale changes across a codebase rather than single file autocomplete. It reads project structure, dependencies, and patterns, applies edits across many files from one prompt, and runs the project's tests to confirm nothing breaks, gating completion on those tests. It lives in the terminal, installs through npm, and works with any tech stack.
A knowledge.md file scaffolded by the /init command gives the agents persistent project context, @filename mentions pin specific files, @AgentName mentions invoke a particular agent, and /history browses and resumes past conversations. The team publishes its own evaluation suite of more than 175 coding tasks and reports Codebuff beating Claude Code on it, a figure worth reading with the usual skepticism given it is self reported against the vendor's own harness.
Three strengths stand out for developers. Model choice: Codebuff runs any model available through OpenRouter, and because the model is a field on each agent definition, different roles in the same workflow can run on different models.
Extensibility: a TypeScript SDK exposes a CodebuffClient whose run method takes an agent id, a prompt, custom agent definitions, custom tool definitions and an event callback, so teams embed the agents in their own applications and CI/CD.
And reuse: a public Agent Store at codebuff.com/store publishes agents that others compose into their own workflows, with custom agents authored as TypeScript definitions declaring their model, their permitted tools and the subagents they may spawn.
For onboarding, Codebuff offers Freebuff, a free ad supported version with no subscription, credits, configuration, or API key, powered by open weight models including DeepSeek, MiMo and MiniMax and monetized through text ads in the command line. Buyers should read its privacy terms: the vendor states that prompts and messages, including pasted content, may be analyzed to personalize those ads, though connected repositories and separate uploads are excluded from advertising providers.
The main caveats are that it is terminal only with no IDE, web or chat surface, that custom agents require TypeScript, that no security attestation, SSO or governance surface is documented, and that despite the open source license the agent authenticates against a Codebuff issued API key and reaches models through OpenRouter, so prompts and code context leave the machine in normal operation.
Vendor details
Canonical URL
https://www.codebuff.com
Category
Coding agent
Subcategory
Open source multi-agent terminal coding agent
Funding status
Independent. Codebuff, Inc., founded by James Grugett and Brandon Chen, based in San Francisco and backed by Y Combinator in the Fall 2024 batch. Launched originally as Manicode in 2024 and renamed to Codebuff following a trademark dispute. Open source under Apache-2.0; the public repository carries roughly 8.3k stars, 983 forks and over 8,000 commits as of August 2026, alongside a commercial managed offering and the free ad-supported Freebuff variant.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Tool calling is deep within the developer's environment: a documented run_terminal_command tool, file reading tools, git analysis and commit workflows, project test execution, bash mode invoked with a leading exclamation mark or /bash, and built-in web research and browser use. MCP servers added in .agents/mcp.json over stdio, HTTP or SSE become tools for all base agents, with Notion, GitHub and remote API examples documented. The TypeScript SDK accepts customToolDefinitions so embedders add their own tools, and agents embed into applications and CI/CD pipelines through it. A public Agent Store at codebuff.com/store publishes reusable agents. Model access is brokered through OpenRouter, with model selection set per agent definition.
In practice
You need a change touching many files, endpoints, and tests at once. Codebuff's agents map the codebase, make the edits from one prompt, and run your tests to verify nothing broke.
You want to try an autonomous coding agent without paying for one. Freebuff is free and ad supported with a daily allowance across several models, and the Apache-2.0 source lets you inspect how the agents work.
You want coding assistance inside your own product or CI pipeline. Codebuff's TypeScript SDK lets you embed its agents, define custom tools, and route each task to whichever model fits the job.
Sources & related URLs
Related / legacy domains
Agentic Index coverage score
9.0 / 14 capabilities · 64%
| Integrations & Tool Calling | Full |
|---|---|
|
MCP servers added in.agents/mcp.json over stdio, HTTP or SSE become tools for all base agents, with Notion, GitHub and remote API examples, alongside a bash agent for terminal commands, a researcher for web search and custom tool definitions through the SDK. Sourcecodebuff.com/docs/tips/mcp-servers and docs/agents/overviewread 2026-09-29 |
|
| Workflow Orchestration | Full |
|
A base agent orchestrates eight specialized agents (editor, reviewer, thinker, researcher, file picker, basher, code searcher), and agents spawn others listed in their spawnableAgents configuration when their spawnerPrompt applies, so work flows between multiple agents. Sourcecodebuff.com/docs/agents/overviewread 2026-09-29 |
|
| Knowledge Grounding & RAG | Partial |
|
File picker and code searcher agents map the relevant parts of a project before editing, and knowledge.md files carry conventions and commands. Context is assembled per run, and no persistent index, graph or embeddings layer is documented. Sourcecodebuff.com/docs/agents/overview, docs/tips/knowledge-files and github.com/CodebuffAI/codebuffread 2026-09-29 |
|
| Human Oversight & Guardrails | Partial |
|
Each agent definition limits the tools it may use and which agents it may spawn, a customer set constraint on what the agent can do, and a reviewer agent checks changes; no step where a person approves an action before it commits is documented on the agents overview. Sourcecodebuff.com/docs/agents/overviewread 2026-09-29 |
|
| Security, Identity & Governance | Unable to verify |
|
No certification, SSO, role based access or audit log is documented; the privacy page says prompts, messages, agent traces, code, files and repository data are used to provide Freebuff, with training use only where stated, and points to a separate privacy policy for retention and advertising choices. Sourcecodebuff.com/docs/advanced/privacyread 2026-09-29 |
|
| Observability & Auditability | Partial |
|
The SDK's event callback streams agent start and finish, tool call and tool result, text and error events as a run executes, so an embedder can watch each step, but keeping that record is left to the embedder and no audit log, retained trace or export is offered. Sourcecodebuff.com/docs/advanced/sdkread 2026-09-29 |
|
| Memory & State Persistence | Partial |
|
The SDK's previousRun parameter carries conversation state from one run into the next, and knowledge.md files, which the user writes or asks /init to generate, load every run as project context; no memory layer the agent writes with a stated scope and lifetime is documented. Sourcecodebuff.com/docs/advanced/sdk and docs/tips/knowledge-filesread 2026-09-29 |
|
| Deployment & Data Residency | Partial |
|
The CLI runs in the developer's terminal and the Apache-2.0 source can be built and inspected, but the agent authenticates with a Codebuff API key and reaches hosted models, so prompts and code leave the machine in normal use; no customer hosted deployment or region choice is documented. Sourcegithub.com/CodebuffAI/codebuff and codebuff.com/docs/advanced/sdkread 2026-09-29 |
|
| Prebuilt Agents, Templates & Packs | Full |
|
A public Agent Store at codebuff.com/store lists reusable agents to browse and adopt, and eight named built in agents each do their own job (editor, reviewer, thinker, researcher, file picker, basher, code searcher and the base orchestrator), each a whole unit on its own. Sourcecodebuff.com/docs/agents/overviewread 2026-09-29 |
|
| Triggers & Channel Coverage | Partial |
|
Agents start from the terminal CLI, by @mention inside a session, or from the customer's own code through the SDK, which the docs position for CI/CD and automation scripts; no schedule, webhook or event trigger that wakes an agent on its own is documented. Sourcecodebuff.com/docs/advanced/sdk and github.com/CodebuffAI/codebuffread 2026-09-29 |
|
| Model Flexibility & Routing | Full |
|
Model choice is a field on each agent definition, so agents in one workflow can run on different models, and Freebuff lets users spend its daily allowance across a list of models including GLM, DeepSeek and GPT models. Sourcecodebuff.com/docs/agents/overview and freebuff.comread 2026-09-29 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
The @codebuff/sdk package exposes a CodebuffClient with an API key and working directory and a run method taking an agent, a prompt, previous run state, custom tool definitions and an event callback, documented for CI/CD, batch jobs, editor extensions and web apps; custom agents are authored in TypeScript and shared through the Agent Store. Sourcecodebuff.com/docs/advanced/sdkread 2026-09-29 |
|
| Testing, Debugging & Optimization | Partial |
|
A reviewer agent checks changes and the repository carries an evals folder, but no page documents how a customer runs evaluations against its own agents or workload, so the harness is not shown as a customer surface. Sourcecodebuff.com/docs/agents/overview and github.com/CodebuffAI/codebuffread 2026-09-29 |
|
| Browser & Computer Use | Partial |
|
The repository states agents can investigate documentation and test applications in a real browser, but no page describes what the agent can do in it, so the capability is asserted rather than documented. Sourcegithub.com/CodebuffAI/codebuffread 2026-09-29 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Unable to verify against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Freebuff free and ad supported; Codebuff from $100 per month, or pay as you go at $0.01 per credit
Credits spent by task complexity, bought through monthly subscriptions ($100, $200 or $500) or pay as you go at $0.01 each; Freebuff is free and ad supported.
Included quota
Freebuff: 100 Freebucks a day. Codebuff subscriptions: $100 for 1x usage, $200 for 2.5x and $500 for 7x.
What is public
Subscription prices, their usage multiples and the $0.01 credit price are published on codebuff.com/pricing, and Freebuff's free daily allowance on freebuff.com; enterprise pricing is not published.
Billing mechanics
Credits are spent by task complexity; subscriptions bundle usage at 1x, 2.5x or 7x, and pay as you go credits cost $0.01 each without a subscription. Freebuff is funded by ads.
Cost watchouts
Multi agent runs consume more credits per task, and the largest subscription reaches $500 a month.
Variable cost rationale
Credits are metered by task complexity and multi agent runs use more per task, so spend tracks usage rather than a flat seat.
Additional watchouts
Multi agent runs spend more credits per task than single agent work, and Freebuff uses prompts, messages, agent traces, code and repository data to provide its service, which is funded by advertising.
Overage / add-ons
Credits beyond a subscription's usage are bought pay as you go at $0.01 per credit; simple queries cost less and complex changes more.
Sales call required
No, self serve available
Free / trial
Free through Freebuff, ad supported, with 100 Freebucks a day and no API key or credit card.
Lowest paid plan
Pay as you go at $0.01 per credit with no subscription; subscriptions from $100 a month.
Commercial notes
Three paths: the free ad supported Freebuff, paid Codebuff subscriptions or pay as you go credits, and the Apache-2.0 source for teams that build the harness themselves.
Key ambiguities
The credit volume behind each subscription's usage multiple is not stated, so cost per task is hard to predict in advance.
Cancellation / refund
Subscriptions can be canceled anytime; Freebuff and pay as you go need no commitment.
Support SLA / resale
Community support for the open source project; enterprise plans are available on request.
Missing data
How many credits a subscription's 1x usage includes, and enterprise pricing, are not published.
Related vendors
- Cognition — Maker of Devin, an autonomous AI software engineer
- 10Web — Agentic website platform whose specialized AI agents build, host,…
- AgentUI — Managed AI app builder for operations teams that generates and hosts…
- Aider — Open source, model agnostic terminal coding agent that edits your…
- Anthropic Claude Code — Anthropic's agentic coding system across terminal, desktop, IDE, web…
- AppFactor — Agentic platform that runs persistent agents across an enterprise…
Alternatives to Codebuff
The closest documented capability profiles to Codebuff among coding agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Blink9.5 / 14Adds documented Security, Identity & Governance
- Tembo11.0 / 14Adds documented Security, Identity & Governance
- Zed10.0 / 14Adds documented Security, Identity & Governance
- 10Web8.5 / 14Adds documented Security, Identity & Governance
- JetBrains AI10.5 / 14Adds documented Security, Identity & Governance
- Potpie AI10.5 / 14Adds documented Security, Identity & Governance
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded