How Anthropic and OpenAI treat your work
What Anthropic and OpenAI say, in their own words, about the code and prompts Limn sends on your account.
When Limn runs a review, a fix or a summary, it sends the work to the AI provider you chose, on your account: the code and the diff, the pull request’s text and comments, and where relevant the ticket and sections of your docs. Nothing goes to a Limn server, because there is none. What happens to that work at the provider is set by the provider’s terms for your kind of account. This page quotes those terms word for word, with a link to each source, so you can decide for yourself.
#At a glance
| How you sign in | Training on your work | How long it is kept |
|---|---|---|
| Claude Code login on Free, Pro, Max or Ultra | Only when your training setting is on. Feedback and content flagged for safety review can be used even when it is off. | 5 years with the setting on; 30 days with it off |
| Anthropic API key, or Claude Team or Enterprise | No, unless your organisation chooses to share data | 30 days as standard |
| ChatGPT Business, Enterprise or Edu | Not by default | Your ChatGPT workspace settings |
| ChatGPT Free, Go, Plus or Pro | Not covered by OpenAI’s “no training by default” line. Check your ChatGPT data controls. | Your ChatGPT settings |
The rest of the page gives the exact words behind each line. For which account Limn uses, see Which account Limn uses.
#Anthropic (Claude)
Which terms apply
Anthropic has consumer terms for its Free, Pro and Max plans, and commercial terms for API keys and business plans. A Claude Code login on a personal plan is consumer use:
These updates apply to users on our Claude Free, Pro, and Max plans, including when they use Claude Code from accounts associated with those plans. They do not apply to services under our Commercial Terms, including Claude for Work, Claude for Government, Claude for Education, or API use, including via third parties such as Amazon Bedrock and Google Cloud's Vertex AI.
Training on a personal plan
We give you the choice to allow your data to be used to improve future Claude models. We will train new models using data from Free, Pro, Max, and Ultra accounts when this setting is on (including when you use Claude Code from these accounts).
We may use Materials to provide, maintain, and improve the Services, including training our models, unless you opt out of training through your account settings. Even if you opt out, we will use Materials for model training when: (1) you provide Feedback to us regarding any Materials, or (2) your Materials are flagged for safety review to improve our ability to detect harmful content, enforce our policies, or advance our safety research.
You're always in control of this setting and whether we use your data in this way.
Privacy settings can be changed at any time at claude.ai/settings/data-privacy-controls
Check the setting at claude.ai/settings/data-privacy-controls before you use Limn on code you would not want used for training.
Training on an API key or a business plan
Anthropic does not train generative models using code or prompts sent to Claude Code under commercial terms, unless the customer has chosen to provide their data to us for model improvement
Anthropic may not train models on Customer Content from Services.
How long Anthropic keeps it
For personal plans, by training setting:
Users who allow data use for model improvement: 5-year retention period to support model development and safety improvements
Users who don't allow data use for model improvement: 30-day retention period
For commercial users (Team, Enterprise and API), the page lists:
Standard: 30-day retention period
One exception applies to everyone:
Transcripts shared via /feedback, or via /bug and /share, which report through the same path, are retained for 5 years.
Limn never sends feedback or bug reports to Anthropic on your behalf.
Who owns the work
As between you and Anthropic, and to the extent permitted by applicable law, you retain any right, title, and interest that you have in such Inputs. Subject to your compliance with our Terms, we assign to you all our right, title, and interest (if any) in Outputs.
Anthropic agrees that Customer (a) retains all rights to its Inputs, and (b) owns its Outputs.
In transit, and sale of data
Claude Code runs locally. To interact with the LLM, Claude Code sends data over the network. This data includes all user prompts and model outputs, encrypted in transit via TLS 1.2+.
We do not sell users' data to third parties.
#OpenAI (Codex)
Your sign-in decides the policy
Limn runs Codex only on a ChatGPT sign-in. OpenAI says:
When you sign in with ChatGPT, Codex usage follows your ChatGPT workspace permissions, role-based access control (RBAC), and ChatGPT Enterprise retention and residency settings.
ChatGPT Work and Codex are included in your ChatGPT Free, Go, Plus, Pro, Business, Edu, or Enterprise plan
Training
OpenAI’s Codex pricing page lists this line in its comparison of plans. Check the page for the plans it covers:
No training on API or business data by default
No training on your business data by default.
Business, Enterprise, and Edu workspace data processed by covered OpenAI services is encrypted in transit and at rest and is not used to train OpenAI models by default.
On a personal ChatGPT plan (Free, Go, Plus or Pro), that line does not apply. Check the data controls in your ChatGPT settings before you use Codex in Limn on code you would not want used for training.
API keys, for context
Limn does not run Codex on an OpenAI API key today. For comparison, OpenAI says an API key follows different settings:
With an API key, usage follows your API organization's retention and data-sharing settings instead.
Your data is your data. As of March 1, 2023, data sent to the OpenAI API is not used to train or improve OpenAI models (unless you explicitly opt in to share data with us).
By default, abuse monitoring logs are generated for all API feature usage and retained for up to 30 days, unless longer retention is required by law, or is reasonably necessary to protect our services or any third party from harm.
#Using your own Claude login
Limn is a program you run on your own computer. When it uses Claude, it uses the Claude Code login you made yourself, through Anthropic’s own sign-in, for your own work. Limn never sees, stores or forwards that login, there is no Limn server in between, and no one else’s requests go through your account.
Anthropic’s policy recommends API keys for products built on its Agent SDK, which Limn uses:
Developers building products or services that interact with Claude's capabilities, including those using the Agent SDK, should use API key authentication through Claude Console or a supported cloud provider. Anthropic does not permit third-party developers to offer Claude.ai login into their own applications, or to route requests through Free, Pro, Max, or Ultra plan credentials on behalf of their users.
Limn offers no Claude sign-in of its own and routes nothing on anyone’s behalf: each person runs Limn with their own access. Anthropic’s plan limits assume ordinary, individual use, and its consumer terms limit automated access outside an API key (Consumer Terms of Service, Anthropic, effective 8 October 2025 (retrieved 2026-11-09)).
#Caveats
- “Not trained on by default” applies to Anthropic only under commercial terms (an API key, Team or Enterprise), and to OpenAI only for business plans and API keys.
- Zero data retention is something an organisation arranges with the provider. It is not the standard.
- OpenAI’s help centre and policy pages could not be quoted here. Read them yourself at openai.com/policies.
What Limn itself keeps, and where, is on What Limn stores.