Go from a task to a production deploy.
Orchestrate a fleet of isolated AI agents from one dashboard. cyql automates the path from a plain-language task to production — plan, code, review, ship — while you shape the workflow with approval gates, guardrails, and model routing. On a durable loop you can audit.
One dashboard to launch, approve, and track every agent run.
Tasks
Add rate limiting to the public API
PlanningImplement dark mode toggle
Awaiting approvalMigrate auth to JWT refresh tokens
ReviewingUpgrade to React 19
Shippedcyql ≈ cycle
One loop, from request to merge
Every task runs the same five-stage loop — and replans itself whenever something needs another pass.
- 01
Describe the task
Write what you want in plain language, link an issue, or send it from Slack. cyql reads your repo, conventions, and past work to ground the request.
- 02
Plan & decompose
The planner breaks the task into a dependency-aware DAG of subtasks, surfaces key decisions and risks, and waits for your approval when it matters.
- 03
Dispatch the fleet
The orchestrator schedules one isolated pod per subtask — scaling out and back in with the queue. No shared state, no blast radius.
- 04
Code & self-review
Agents write code, run your tests and linters, and review each other's diffs against your standards. Failures route back to a replan, not to you.
- 05
Ship a clean PR
You get a focused, reviewed pull request with a written rationale and a full execution trace. Approve, request changes, or let it auto-merge.
Platform
Everything an agent needs to ship safely
A durable workflow engine, hardened isolation, and the integrations your team already uses.
The plan → ship loop
A durable workflow engine drives every task from draft to merged PR, resuming exactly where it left off after any failure.
Runtime isolation
Each agent runs in a locked-down, isolated pod with egress firewalled to your allow-list. Cloud metadata and private ranges are blocked by default.
Parallel agents
Independent subtasks run at the same time across an autoscaling worker pool, so big changes land in minutes, not hours.
Built-in review
Every diff is checked against your conventions and test suite before it reaches you — agents review agents, you review the result.
Short-lived credentials
Workers never hold long-lived secrets. The orchestrator mints per-task, scope-limited tokens that expire when the job ends.
Full execution traces
Replay every decision, prompt, tool call, and file change. Cluster recurring failures and turn them into guardrails.
Security by design
Autonomy your security team can say yes to
Every agent runs in an isolated pod with firewalled egress and no long-lived secrets. The metadata endpoint and private ranges are blocked by default.
Pod-per-task isolation
No shared state, no blast radius between tasks.
Short-lived credentials
Scoped tokens minted per task, expired on exit.
Loved by builders
Engineers ship more with cyql
“cyql cleared a backlog of dependency bumps and flaky-test fixes in a weekend. The PRs were small, well-described, and passed review like a senior engineer wrote them.”
“The isolation model was what got it past our security team. Per-task credentials and locked-down egress meant we could say yes without a six-month review.”
“We point cyql at a Linear ticket and a clean PR shows up. The execution trace means I always know exactly what it did and why.”
FAQ
Questions, answered
Everything you need to know before your first run.
How is cyql different from an AI coding assistant?
Assistants help you write code in your editor. cyql owns the whole task: it plans, dispatches a fleet of agents, writes and reviews the code in isolated runtimes, and opens a finished pull request. You stay in the loop on approvals — not on keystrokes.
Does cyql have access to my code and secrets?
Agents run in locked-down pods with no long-lived secrets. The orchestrator issues short-lived, scope-limited credentials per task, and egress is restricted to an allow-list. On Enterprise you can self-host so code never leaves your network.
Which languages and stacks are supported?
cyql is language-agnostic — it works in your repo with your tools, running your existing tests, linters, and build. Teams use it across TypeScript, Python, Go, Rust, Java, and more.
What happens when an agent gets it wrong?
Failures route back into a replan rather than to your inbox. Every run produces a full trace you can replay, and recurring failures are clustered so you can add a guardrail once and prevent them for good.
Ship your backlog while you sleep.
Start free in minutes. Connect a repo, describe a task, and watch a reviewed pull request show up.
No credit card · Free forever for individuals