FAQ
Short answers, each linking to the page that covers it properly. If you are stuck on a specific symptom, common issues is indexed that way instead.
Getting in
Section titled “Getting in”How do I get access?
Section titled “How do I get access?”CommandChain is in a closed beta. Apply from the product page, and you will get an invitation by email once your application is reviewed. See the quickstart for what happens after that.
What it does
Section titled “What it does”What actually is CommandChain?
Section titled “What actually is CommandChain?”You set direction, and a team of AI agents shapes, builds, launches, and grows the work, while you keep the decisions that matter. See what is CommandChain.
What does an agent do without me watching?
Section titled “What does an agent do without me watching?”Exactly as much as you have allowed. Every class of action sits on an autonomy ladder from “ask me each time” through “act, then show me” to “act within a policy”, and some classes are locked so they always need a person. See approvals and the glossary.
Do I need to know how to code?
Section titled “Do I need to know how to code?”No. Coding work is one of the things the platform does, not the way you use it.
Code and data
Section titled “Code and data”Where does my code go?
Section titled “Where does my code go?”Onto a machine you choose. A runner checks the repository out, runs the coding agent against it, commits, and pushes. On a self-hosted runner that machine is yours, and your source never leaves it. See workspaces and git.
Where does my data live?
Section titled “Where does my data live?”Your workspace (huddles, tasks, runs, projects, and the audit log) lives in the hosted control plane. Your source code stays on the runner that does the work. See workspaces and git.
Can an agent read the rest of my machine?
Section titled “Can an agent read the rest of my machine?”No. Every agent process runs inside an operating-system boundary the runner builds before it starts, with a synthetic home directory holding only the sign-in for that one attempt. If a machine cannot enforce that boundary, the runner refuses to run an agent there at all. See the runner security model.
What happens to my coding-agent login?
Section titled “What happens to my coding-agent login?”On your own machine, nothing. The runner keeps its own sign-in there and never reads, changes or uploads yours. See sign-in only or full setup.
On managed servers, runs use a login you upload from the runner screen (press c) or with creds push. It is stored encrypted and released to one run attempt at a time. See managed runner credentials.
Models and intelligence
Section titled “Models and intelligence”Which AI models do agents use?
Section titled “Which AI models do agents use?”Whichever you allow. A workspace sets a default model and an allowlist, and each agent can be pinned to something specific. See the model catalog.
Do I need an API key to start?
Section titled “Do I need an API key to start?”No. New workspaces get trial credits, and after that agents think through a coding agent on your runner, which the installer signs in with your browser. Prefer a key? Press a in Sign-ins (s) on the runner screen. See install a runner.
What happens when trial credits run out?
Section titled “What happens when trial credits run out?”The request stops before it calls a model and returns a 402 (trial_credits_exhausted). Nothing is charged. Pair a runner to keep going. See trial credits.
Marketing and publishing
Section titled “Marketing and publishing”Can an agent publish something without asking me?
Section titled “Can an agent publish something without asking me?”Not by default. Every externally visible action starts at “ask me each time”, so a publish becomes an approval carrying the exact payload that will be sent. See autonomy for marketing actions.
Can an agent spend money on advertising?
Section titled “Can an agent spend money on advertising?”No. ad.spend is locked at “ask me each time” in code, not in configuration, and nothing raises it. Agents can plan a paid campaign and draft the creative; no money moves. See the marketing loop.
Do I have to connect anything to use marketing?
Section titled “Do I have to connect anything to use marketing?”No. Drafting, atomizing, scheduling, and the calendar work from the first day. Only publishing needs an integration, and you can skip that: post from your own account and type the numbers back in. See publish by hand.
Do I need a developer account to post to X or Threads?
Section titled “Do I need a developer account to post to X or Threads?”Yes, for both, and X charges $0.20 per post carrying a link. Threads is free, but only invited testers can connect until your Meta app passes review. See connect social platforms.
Runners
Section titled “Runners”Do I have to install a runner?
Section titled “Do I have to install a runner?”For real work, yes. Anything that touches a repository, a filesystem, or a command line happens on a runner. Conversation-only workspaces can run on Direct-LLM without one.
How do I see what my runner needs?
Section titled “How do I see what my runner needs?”Run commandchain-runner on that machine and follow its prompt. See the runner screen.
What does the installer put on my machine?
Section titled “What does the installer put on my machine?”One executable in ~/.local/bin, a 0700 directory at ~/.commandchain, and a log directory. By default there is no service at all: the runner runs in the terminal you start it in. Install with --service and you also get a user-level service. No administrator rights, no runtime, no package manager, and no token to paste. See install a runner.
Can I run a runner on Windows?
Section titled “Can I run a runner on Windows?”Not a self-hosted one, because there is no proven unprivileged Windows sandbox. Run on CommandChain’s managed servers instead. See connect an AI account for managed runs.
Setup and limits
Section titled “Setup and limits”Why did my invitation email never arrive?
Section titled “Why did my invitation email never arrive?”Check spam first. If it is not there, the invitation record still exists: whoever sent it can copy the link from the invite dialog and send it to you directly, or re-send. See users and invitations.
Can I approve things from Slack?
Section titled “Can I approve things from Slack?”You can receive notifications and use slash commands in Slack. Approvals happen in the web app.
Next steps
Section titled “Next steps”- Common issues: symptom-first fixes.
- Get help: where to reach us, and what to include.
- Glossary: every term, defined once.