- What is an AI coding agent?
- An AI coding agent is software that takes a development task and carries it out on a codebase by itself: it reads the code, makes the change, runs commands and tests, and hands back the result. CodeLynq is an AI coding agent for teams: it works from a spec your team approved, checks its own work against your tests, a self-review, a security scan and your architecture rules, and ends in a pull request a person reviews and merges.
- How is CodeLynq different from GitHub Copilot, Cursor or Devin?
- Copilot and Cursor make a developer faster while they write code; CodeLynq takes work no developer has picked up yet and turns it into a pull request without anyone at the keyboard. Devin also works on its own; CodeLynq differs by requiring an approved spec before any code is written, letting you choose the coding agent (Claude Code, OpenAI Codex or Grok), and offering runners on your own hardware. Our comparison pages go into each one.
- How do you stop an AI agent from eroding our architecture?
- CodeLynq maps the structure of your repositories from the code and lets you write down which parts may not depend on which. Every run is checked against those rules before a pull request opens, and a violation goes back to the agent to fix. Each pull request lists what the change did to the structure, so architecture drift is visible instead of buried in a diff.
- Do you help us roll CodeLynq out across our teams?
- Yes. Beyond the product, we help you get CodeLynq working at your company: connecting repositories, writing down your conventions, running a guided pilot, training your engineers and scaling it from there. Reach out and we will tailor an onboarding plan to your stack and workflow.
- Does work have to come from a meeting?
- No. A meeting note or transcript is one way in, not the only one. A two-line written request works. So does a production alert from your monitoring, a failing CI run on an open pull request, an item picked up during AI backlog grooming, or another AI assistant calling CodeLynq over MCP. Everything lands in the same reviewed pipeline.
- What does our security team need to know?
- Single sign-on over SAML 2.0 or OIDC, with password login blocked for your domain once you enforce it. SCIM 2.0 provisioning and deprovisioning from your IdP, organisation-wide enforced MFA, CIDR IP allowlists and admin/member roles. Every privileged action lands in an audit log you can filter, export to CSV, or stream as JSON to an HTTPS endpoint you control — and you set how long audit and run logs are kept. Repository tokens and model keys are encrypted at rest with Fernet, and a run only ever receives the credentials it needs.
- Does our source code have to leave our infrastructure?
- Not if you do not want it to. A self-hosted runner executes runs on your own machines: repositories are cloned locally and never reach our servers, while specs, approvals, comments and the audit trail stay in CodeLynq. You can also bring your own model key, which puts the spend and the provider’s data-retention terms under your own contract.
- Does CodeLynq write code directly to my main branch?
- No. Every development run ends in a pull request in your connected repository. Your team reviews and merges through its normal workflow. CodeLynq never pushes to protected branches.
- Which AI model writes the code? Can we choose?
- You choose the agent per organisation: Claude Code, Grok, or the OpenAI Codex CLI. Whichever you pick, the pipeline around it is the same — the same approved spec goes in, and the result still passes your verify command, a self-review of the diff and a security scan before a pull request opens. The agent is a component rather than the product, which is also why switching is a setting rather than a migration. For Claude Code you can additionally pin a model, or let CodeLynq route per task so a dependency bump does not run on the most expensive option.
- What is a development run?
- One autonomous implementation of an approved Technical Solution Document. The run reads your repository, writes the change, then checks its own work — your verify command, a self-review of the diff, a security scan and a docs pass — fixing what it finds before opening a pull request. If CI then fails or a reviewer requests changes, it can iterate on the same branch. Plans include a monthly run allowance you can track in organization settings.
- How does CodeLynq understand my codebase?
- You connect a repository through the guided wizard, and CodeLynq grounds both the specs it drafts and the code it writes in what is actually there. It also keeps a durable memory per project: facts learned from previous runs and from your reviewers’ feedback — conventions, where things live, pitfalls — so later runs start informed rather than from scratch.
- Who stays in control of what gets built?
- You do, at every handover. Tasks are validated in chat before anything is specified, the TSD needs your explicit approval before a line of code is written, and the pull request goes through your normal review. CodeLynq never merges anything itself and never pushes to a protected branch. Anything that runs automatically — iterating on CI feedback, acting on an alert, triggering a deploy — is off by default and switched on per project.
- Can I change or cancel my plan?
- Yes. You can upgrade, downgrade or cancel from your organization’s billing settings at any time. Every plan comes with a 14-day money-back guarantee. If it is not for you, ask for a full refund within 14 days.