Bosun is live · free · last verified 23 Oct 2026 · viability monitor
Bosun
Open-source control plane that routes coding tasks across multiple AI executors and owns the PR lifecycle from issue to merged commit.
Bosun is a genuine control plane, not a chat wrapper — weighted executor routing, executor failover, circuit breakers, stale-claim reclamation, and decisioned PR gates are real orchestration engineering. It earns its place for teams already running GitHub Actions or similar CI who want to supervise several coding agents without duplicate work. It differs from IDE-bound agents like GitHub Copilot or an OpenAI Codex setup because the unit of work is the pull request, not the autocomplete suggestion: Bosun opens the branch, watches CI, rebases on conflict, and merges when gates pass. The tradeoff is total self-hosting — Node.js 18+, git on PATH, and a .env you write yourself, plus docs that
Verdict generatedVerified 7d ago · liveness 68/100 · cite: rightaichoice.com/tools/bosun
- Engineering teams automating the PR lifecycle from branch to merge
- Teams running multi-agent fleets across Copilot, Codex, and Claude
- Organizations that need circuit breakers and failover in agent workflows
- Advanced users comfortable with CLI, Node.js, and Docker
- Non-technical teams looking for a no-setup SaaS
- People who want a single-agent ask-and-answer assistant like ChatGPT
- Teams unwilling to manage self-hosted infrastructure (Node.js 18+ and git required)
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Bosun if you want an assistant inside your editor rather than a supervisor over your pull requests — it assumes you already run CI/CD and are willing to write a .env, install Node.js 18+ and git, and operate the fleet yourself.
Self-hosting is the cost: you pay for the machines or containers the agents run on, plus whatever your executor providers charge per token — Bosun adds no per-seat fee but does not absorb any of that.
Bosun is published as free and open source, so there is no per-seat or per-task bill from the project itself. That puts it below commercial agent-orchestration platforms on license cost and above them on operational cost: you supply the compute, the executor API spend, and the operator time. For a small team already paying for Copilot, Codex, or Claude seats, Bosun's cost line is infrastructure plus those subscriptions you already have, not a new subscription.
In short
Bosun — Open-source control plane that routes coding tasks across multiple AI executors and owns the PR lifecycle from issue to merged commit. Best for Engineering teams automating the PR lifecycle from branch to merge, Teams running multi-agent fleets across Copilot, Codex, and Claude, Organizations that need circuit breakers and failover in agent workflows. Free to use.
What is Bosun?
Bosun — Open-source control plane that routes coding tasks across multiple AI executors and owns the PR lifecycle from issue to merged commit. Best for Engineering teams automating the PR lifecycle from branch to merge, Teams running multi-agent fleets across Copilot, Codex, and Claude, Organizations that need circuit breakers and failover in agent workflows. Free to use.
Bosun is a self-hosted, open-source control plane for autonomous software engineering. You install it with `npm install -g bosun`, run the interactive setup wizard (`bosun --setup`) in Recommended mode or write a single `.env`, then launch `bosun` — the supervisor auto-connects to your configured board and executor pool. Routing is weighted across executor backends: its own configuration examples show Copilot on Claude Opus 4.6 at 50% weight alongside Codex at 50%, with Claude, GPT, Gemini, and OpenCode also listed as executor integrations. When an executor hits a cooldown or retry limit, Bosun fails the task over to another available agent instead of stalling. The PR lifecycle is the product's core: Bosun creates the PR, watches CI, auto-rebases on conflicts, applies decisioned gates that combine CI status with review requirements, and merges when checks pass. A self-healing layer detects failure patterns via autofix, uses circuit breakers to stop error loops, and reclaims stale claims through heartbeat liveness monitoring so two agents don't duplicate work on the same issue. Task intake works against a built-in internal board plus native adapters for GitHub Issues, GitHub Projects v2, and Jira with bidirectional sync. You operate it from an interactive CLI shell, a Telegram bot with 30+ commands and a 7-tab Mini App (Dashboard, Tasks, Agents, Infra, Control, Logs, Settings), or an Electron desktop app, with optional WhatsApp notifications and voice mode. Agents can be isolated in Docker, Podman, or Apple Container sandboxes, and fleet coordination tracks presence, sessions, and workspaces across multiple machines. This is infrastructure for engineering teams that already run CI/CD and want a safety layer over a fleet of coding agents — not a turnkey SaaS.
Key features
- Multi-executor routing with weighted distribution across Copilot, Codex, Claude, Gemini, and OpenCode
- Automatic failover when an executor hits a cooldown or retry limit
- Smart PR lifecycle: create, monitor CI, auto-rebase on conflict, merge when checks pass
- Decisioned PR gates combining CI status and review requirements into explicit merge/no-merge outcomes
- Automated pass/fail labeling on agent-authored pull requests
- Self-healing recovery: autofix failure-pattern detection and circuit breakers to prevent error loops
- Stale claim reclamation via heartbeat liveness monitoring
- Atomic task claims preventing duplicate work across workstations
- Task boards: internal board plus GitHub Issues, GitHub Projects v2, and Jira with bidirectional sync
- Telegram bot with 30+ commands, inline keyboards, live digests, and a 7-tab Mini App
- Mini App tabs: Dashboard, Tasks, Agents, Infra, Control, Logs, Settings
- Electron desktop app
- Interactive CLI shell mode and Sentinel watchdog companion
- Optional WhatsApp notification channel
- Optional voice mode control surface
Behind the verdict
Bosun sits in an unusual spot: it is not an agent, it is the layer that supervises agents. That distinction is the whole argument for it. If you have tried to let a coding agent own a repository unattended, you already know the failure modes — the run stalls, the executor rate-limits, two agents pick up the same issue, a PR sits unreviewed for hours, a conflict goes unresolved. Bosun names each of those and assigns a mechanism: failover across the executor pool, atomic claims plus heartbeat liveness with stale sweeps, decisioned PR gates that merge only when CI status and review requirements agree, autofix failure-pattern detection with circuit breakers to stop error loops. That is a coherent design, and it is the reason the seed data describes it as a control plane rather than a wrapper. The routing model deserves specific attention. In Bosun's own configuration examples, the executor pool is declared as a weighted list — `EXECUTORS=COPILOT:CLAUDE_OPUS_4_6:50,CODEX:DEFAULT:50` — so you can spread work across vendors by percentage instead of betting the pipeline on one provider. Bosun's docs list Codex, Copilot, Claude, Gemini, and OpenCode integrations. If one of those vendors has an outage, a price change, or a model regression, you are not rewriting your automation; you reweight the pool. Bosch's failover also fires on cooldown and retry limits, which is the practical failure you hit in production. The PR lifecycle is where Bosun spends most of its engineering. Creating a PR is easy; the hard part is everything after — CI that fails transiently, conflicts that appear while a run is queued, review approvals that gate a merge, labels that must reflect pass or fail for downstream tooling. Bosun treats the merge as a decision with inputs rather than a step that just happens. Teams that already have branch protection and required checks will recognize this shape immediately. The control surfaces are broader than most self-hosted agent tooling bothers with. A Telegram bot with 30+ commands and a 7-tab Mini App means you can watch a fleet from a phone without SSH. The Electron desktop app and interactive CLI shell cover the desk. Optional WhatsApp notifications and optional voice mode are extras, not the core. Container isolation via Docker, Podman, or Apple Container with image management and concurrent container limits is the piece that makes running untrusted generated code less alarming. The honest weaknesses: this is infrastructure you own end to end. You supply the Node.js 18+ runtime, git, GitHub credentials, Telegram bot token, executor API access, and container runtime. The homepage counter block still renders placeholder zeros, and the docs are static pages published from the repo rather than a hosted knowledge base — acceptable for a project under active development, less so if you need a vendor SLA. Note also that the project's public site shows version markers that do not always agree with each other (the homepage banner versus the
Models under the hood
as of 2026-10-22
Limitations
- Bosun is open-source and self-hosted, so you manage the infrastructure, configuration (a written .env, Node.js 18+, and git on PATH), and scaling yourself.
- Docs are static pages published from the repo, and the homepage stats block still renders placeholder zeros, so expect rough edges.
- The public site's version markers do not all agree — the homepage banner showed v0.0.0 while the docs pages state v0.43.2 — so confirm the version you actually install.
- Nothing is managed for you: Telegram credentials, executor API access, and container runtimes are your responsibility.
- Some capabilities are explicitly optional and off by default, including WhatsApp notifications, voice mode, and container isolation, which means enabling them adds configuration and failure surface you own.
as of 2026-10-13
Researching Bosun? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Is Bosun still active?
Bosun is live · free · last verified 23 Oct 2026 · viability monitor.
Viability score
How well maintained and how widely used is Bosun? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →What do users say about Bosun?
What people actually say about Bosun — is it worth it?
We scanned public community sources for Bosun on Sep 9, 2026 and could not establish that the discussion we found is about this tool rather than something else sharing its name. Only 3 of the posts we fetched could be positively tied to Bosun. Rather than publish a sentiment score built on the wrong subject, we publish nothing here and re-run the scan.
How we verified this
Every listed fact about Bosun— pricing, features, links, status — is re-read from the vendor's own pages by our verification pipeline and each change is logged with a date; the last pass ran on 23 Oct 2026. Pages mix model-assisted synthesis with human review; see how we verify.
Verification history
We have re-verified Bosun 11 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — changed limitations, description, hidden costs, setup time, integrations, our verdict, pricing tiers, switching in, pricing fit, models, switching out, features, who should skip it
- — changed platforms, models, pricing page, limitations
- — changed pricing tiers
- — changed hidden costs, description, our verdict, features, who it suits, integrations, limitations, switching in, switching out, models, who should skip it, pricing fit, setup time, summary, use cases
- — changed domain expiry
- — changed SSL certificate
Showing the 6 most recent of 11 verification passes.
Free to cite with attribution — this page re-verifies continuously.
How much does Bosun cost in 2026?
Bosun: Free to use. Pricing last verified 23 Oct 2026.
Plans compared
Every published Bosun tier, its price as the vendor publishes it, who it fits, and what it adds over the tier below. The 12-month cost projection follows.
| Plan | Price | Billing | Who it's for |
|---|---|---|---|
| Open source | Free | — | Engineering teams with existing CI/CD that are willing to self-host and want to supervise multiple coding agents without a per-seat license.Adds: Starting tier: free and open source, with cost shifting to your own infrastructure and executor API spend rather than a Bosun subscription. |
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Where the pricing makes sense
The company stage and team size where Bosun's pricing actually pencils out — and where peers do it cheaper.
Bosun is published as free and open source, so there is no per-seat or per-task bill from the project itself. That puts it below commercial agent-orchestration platforms on license cost and above them on operational cost: you supply the compute, the executor API spend, and the operator time. For a small team already paying for Copilot, Codex, or Claude seats, Bosun's cost line is infrastructure plus those subscriptions you already have, not a new subscription.
Has Bosun's pricing changed?
No plan price for Bosun has moved since we began logging in Aug 2026; the pricing page is re-read on every verification pass.
Who should pick Bosun?
Pick Bosun if you are Engineering teams automating the PR lifecycle from branch to merge, Teams running multi-agent fleets across Copilot, Codex, and Claude, Organizations that need circuit breakers and failover in agent workflows, Advanced users comfortable with CLI, Node.js, and Docker.
Look elsewhere if you are Non-technical teams looking for a no-setup SaaS, People who want a single-agent ask-and-answer assistant like ChatGPT, Teams unwilling to manage self-hosted infrastructure (Node.js 18+ and git required), Beginners unfamiliar with terminals and .env configuration.
Skip it if Skip Bosun if you want an assistant inside your editor rather than a supervisor over your pull requests — it assumes you already run CI/CD and are willing to write a .env, install Node.js 18+ and git, and operate the fleet yourself.
GeneratedReal-world workflow fit
Concrete scenarios for the personas Bosun actually fits — and what changes day-one when you adopt it.
Set GITHUB_REPO, a GitHub token, and an executor pool split 50/50 between Copilot on Claude Opus 4.6 and Codex in a single .env, point KANBAN_BACKEND at a GitHub project board, and run `bosun` so the supervisor picks up labeled issues and dispatches them.
Outcome: Issues flow to branches and PRs without anyone assigning work by hand, and when one executor cools down the task fails over instead of stalling in the queue.
Configure decisioned PR gates so agent-authored PRs merge only when CI checks pass and required reviews are approved, with pass/fail labeling applied automatically, and watch the run state from the Telegram bot's Mini App.
Outcome: Merge decisions become explicit and auditable instead of a reviewer eyeballing a bot's output, and stuck runs surface as alerts rather than silent backlog.
Install Bosun on a workstation and a second box, rely on presence tracking and workspace registry to keep them from claiming the same issue, and isolate agent runs in Docker containers with concurrency limits.
Outcome: Two machines work the same repo without duplicating effort, and generated code executes inside containers rather than directly on the host.
Use cases
- Turn GitHub issues into pull requests end to end, with CI monitoring and merge handled automatically.
- Split coding work across multiple vendors by weight instead of committing to one AI agent.
- Run several AI workstations that share state and avoid duplicate work on the same issue.
- Enforce merge policy on agent-authored PRs using CI status, review approvals, and pass/fail labeling.
- Recover automatically from stalled agent runs using retry policy, cooldowns, and escalation hooks.
- Keep GitHub Projects v2 or Jira in sync bidirectionally with the board your agents actually work from.
- Monitor and steer running agents from a phone via the Telegram bot and Mini App.
- Isolate agent execution in Docker, Podman, or Apple Container sandboxes with concurrency limits.
Setup time & first value
How long it actually takes to get something useful out of Bosun — broken out by persona, not the marketing-page minute.
A single developer who has Node.js 18+ and git ready can reach a first supervised task in roughly 20-40 minutes using `npm install -g bosun` plus the Recommended-mode setup wizard. A team wiring GitHub Projects v2 or Jira boards, a Telegram bot, container isolation, and a weighted multi-executor pool should budget half a day to a day before the first unattended merge.
Switching to or from Bosun
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From GitHub Actions alone: keep your existing workflows and let Bosun drive the PR lifecycle on top of them rather than replacing the CI runner.
- →From a single coding agent in an IDE: move task intake to a GitHub or Jira board, declare that vendor as one entry in the weighted executor pool, and let Bosun open the PR instead of you.
- →From ad-hoc scripts using the GitHub CLI: replace hand-rolled branch and PR scripts with Bosun's task board adapters and decisioned merge gates.
- ↗To GitHub Actions plus manual agent runs: drop the supervisor and keep only the CI workflows you already wrote, accepting that PR creation, rebasing, and stale-run recovery become manual again.
- ↗To a hosted agent-orchestration platform: move task intake and merge policy into a managed service, trading operator control and self-hosting for someone else running the fleet.
Integrations
Resources & guides
- Documentationbosun.engineer
Docs · Bosun
Full product docs from bosun.engineer
- Quickstartbosun.engineer
Getting Started · Bosun
Get up and running fast from bosun.engineer
- Documentationbosun.engineer
Installation · Bosun
Full product docs from bosun.engineer
- Documentationbosun.engineer
Configuration · Bosun
Full product docs from bosun.engineer
- Documentationbosun.engineer
Cli Reference · Bosun
Full product docs from bosun.engineer
- Documentationbosun.engineer
Architecture · Bosun
Full product docs from bosun.engineer
- Documentationbosun.engineer
Integrations · Bosun
Full product docs from bosun.engineer
Tutorials & Learning
YouTube returned 6 videos for “Bosun”, and we withheld 6: 6 could not be judged, because “Bosun” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Bosun.
Official links
What are the best Bosun alternatives?
The closest alternatives to Bosun we list are Refact.ai, Shep and Optio. Ranked by direct product-type match, not category overlap.
| Alternative | Pricing | Starts at | Last verified |
|---|---|---|---|
| Refact.ai | Freemium | Pricing not published | — |
| Shep | Free | Free | — |
| Optio | Free | Free | — |
Alternatives to Bosun
View all
Refact.ai
Open-source autonomous AI coding agent for VS Code and JetBrains that plans, executes, and deploys tasks with step-by-step reasoning.
Tools that pair well with Bosun
Common stack mates teams adopt alongside Bosun, with the specific reason each pairing earns its keep.
Refact.ai
Open-source autonomous AI coding agent for VS Code and JetBrains that plans, executes, and deploys tasks with step-by-step reasoning.
Shep
Open-source, local-first orchestrator that runs parallel AI coding agents in isolated git worktrees, from a one-line idea to a merged pull request.
Optio
Open-source AI coding agent orchestration that turns tickets, crons, and webhooks into merged pull requests on your own Kubernetes cluster.
Bosun vs its closest rival
Bosun's closest head-to-head is with Locus Robotics — Choose Locus Robotics if you need physical warehouse automation to handle high-volume order fulfillment with a proven 2-3x productivity boost. Choose Bosun if you are a development team wanting to automate the entire PR lifecycle using multiple AI coding agents with failover and self-healing. They solve entirely different problems – warehouse logistics vs. software engineering automation. Verdict generated
Head-to-head comparisons
Bosun vs Locus Robotics
Choose Locus Robotics if you need physical warehouse automation to handle high-volume order fulfillment with a proven 2-3x productivity boost. Choose Bosun if you are a development team wanting to automate the entire PR lifecycle using multiple AI coding agents with failover and self-healing. They solve entirely different problems – warehouse logistics vs. software engineering automation.
GeneratedBosun vs Truleo
Bosun and Truleo serve fundamentally different domains. Bosun is ideal for engineering teams wanting a free, open-source control plane to automate AI-assisted PR workflows. Truleo is a paid, CJIS-compliant platform tailored for law enforcement agencies to connect siloed intelligence data. Your choice depends entirely on whether you need autonomous PR lifecycle management or law enforcement intelligence automation.
GeneratedBosun vs Presto Voice
Presto Voice is the clear choice for QSR chains wanting proven drive-thru automation with measurable revenue lift (up to 6% monthly) and high accuracy (95% non-intervention). Bosun is for engineering teams seeking an open-source, multi-agent PR automation tool, but it's early-stage and requires self-hosting. Choose based on domain: restaurant ops vs developer workflow.
GeneratedFrequently asked questions
Used Bosun? Help shape our editorial sentiment research.