How to Fix Cody Sign-In and Access Problems After the Plan Changes

If Cody stopped working or you cannot sign in, the 2025 discontinuation of the Free and Pro plans may be why. Here is how to diagnose access issues and get an iOS-ready setup running again.

Understand What Changed

In 2025, Sourcegraph announced it was discontinuing the standalone Cody Free and Cody Pro plans, and removing Cody from the Enterprise Starter plan.

If your Cody used to work and suddenly does not, this is the first thing to check. It is not necessarily a bug in your setup.

Cody is now positioned mainly as an Enterprise product, and Sourcegraph pointed individual developers toward its newer Amp tool.

Understanding this framing saves hours of local troubleshooting for an access change that no editor setting can fix.

Confirm Whether You Have Current Access

Try signing in to your Sourcegraph account in a browser, separate from the editor. This isolates account status from extension issues.

If your old Free or Pro access ended, you will need access through a current plan, typically Enterprise via your organization.

Check with your team or administrator about whether your company has a Sourcegraph Enterprise instance you should connect to.

If you were a solo user on the old free tier, recognize that the same free path may no longer exist and plan accordingly.

Point the Extension at the Right Instance

A very common error is the editor pointing at the wrong endpoint. Enterprise users must connect to their organization's Sourcegraph instance URL, not a generic default.

Open Cody's settings and verify the server or instance URL matches what your administrator provided.

After correcting the URL, sign out completely and sign back in so the new endpoint takes effect.

A mismatched instance produces authentication failures that look like broken software but are really a configuration issue.

Clear Stale Credentials

Old tokens and cached sessions from a previous plan can block a clean sign-in. Sign out of Cody in the editor to clear the current session.

If a manual access token is configured, remove any expired token and generate a fresh one from your current instance.

Reload the editor window, then start the sign-in flow again from a clean state.

Stale credentials are a frequent cause of persistent authentication errors after a plan or instance change.

Check Network, VPN, and SSO

Enterprise sign-in often runs through single sign-on and corporate networks. If your VPN or SSO session is down, authentication will fail.

Confirm you are on the required network and that your SSO session is active in a browser.

Proxies and firewalls can block the endpoints Cody needs. If logs show network or TLS errors, involve IT with those specifics.

Test from a different network when possible to determine whether the block is environmental rather than account-related.

Read the Error and Logs

Do not guess. Open Cody's output or log channel and read the actual error message, which usually names the cause.

Messages about invalid credentials point to sign-in or token problems. Messages about the instance or endpoint point to a URL misconfiguration.

Rate-limit or plan-related messages indicate an account or usage constraint rather than a broken install.

Matching the log message to the right category prevents you from resetting things that were never the problem.

Decide Your Path Forward

If you are on a team, the cleanest fix is getting proper access through your organization's Sourcegraph Enterprise setup, then connecting the extension to that instance.

If you are an individual who relied on the old free tier, evaluate current options on Sourcegraph's official pricing, including whether Amp fits your workflow.

Compare alternatives too. Several AI assistants integrate with VS Code and JetBrains for Swift work if Cody's new positioning does not fit you.

Whatever you choose, verify pricing and plan details on the official site rather than trusting secondhand figures, since packaging continues to evolve.

Remember the iOS Boundary

Restoring Cody access improves how you read and write Swift, but it changes nothing about how you ship.

You still build, run the simulator, code sign, and submit your app in Xcode. Cody is an assistant, not a build or release tool.

Shipping to the App Store still requires an active Apple Developer Program membership, regardless of which AI assistant you use.

Keep that boundary clear so an access hiccup with an assistant never blocks your actual release pipeline, which lives entirely in Apple's tools.

Migrate a Team Cleanly

If your team is moving from the old individual plans to an Enterprise instance, coordinate the switch so no one is left guessing.

Share the correct instance URL and the expected sign-in method, since a wrong endpoint is the most common failure during migration.

Have each developer sign out fully, clear any stale token, then sign in against the new instance to avoid mixing old and new sessions.

A short written runbook turns a confusing access change into a five-minute step for every engineer.

Evaluate Whether Cody Still Fits

An access change is a natural moment to reassess. If Cody's Enterprise-focused positioning matches your team, connecting to the instance is the clean path forward.

If you are an individual who valued the old free tier, weigh Sourcegraph's newer Amp tool and other assistants that integrate with VS Code and JetBrains for Swift.

Base the decision on your real workflow, not on habit. The tool that lives in your editor and understands your code matters more than brand loyalty.

Whichever you pick, confirm current pricing and terms on the official site before committing budget.

Keep Your Release Path Independent

Whatever you decide about assistants, make sure your ability to ship never depends on any single AI tool being available.

Your build, signing, and submission steps live entirely in Xcode and App Store Connect, so an assistant outage should never block a release.

Keep an active Apple Developer Program membership and a working Xcode setup as the stable core of your pipeline.

With that foundation, swapping or troubleshooting an AI assistant becomes a productivity question, not a shipping emergency.

Document the Working Setup

Once access is restored, write down exactly what worked: the instance URL, the sign-in method, and any token steps.

A short internal note turns the next occurrence into a quick lookup rather than a repeat investigation for you or a teammate.

Include how to tell an access problem apart from a network or SSO problem, since the fixes differ and the log messages point the way.

Capturing this while it is fresh is the cheapest insurance against losing another afternoon to the same access hiccup later, and it helps new teammates get connected on their first day without pinging you for help.

Frequently Asked Questions

Why did Cody suddenly stop working for me?

Sourcegraph discontinued the standalone Cody Free and Pro plans in 2025. If you relied on those tiers, you now need access through a current plan such as Enterprise.

What is the fix if I only ever used the free tier?

Check Sourcegraph's official pricing for current options, get access through your organization's Enterprise instance if available, or evaluate alternatives such as Sourcegraph's newer Amp tool.

Why does sign-in fail even with a valid account?

Often the extension points at the wrong instance URL, or stale tokens and an inactive SSO or VPN session block authentication. Verify the instance URL, clear old credentials, and confirm your network.

Does fixing Cody access help me ship my iOS app?

Only indirectly. Cody helps you write Swift, but building, signing, and submitting to the App Store still require Xcode and an Apple Developer Program membership.