> For the complete documentation index, see [llms.txt](https://docs.otari.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.otari.ai/privacy/agent-session-content.md). # Agent Session Content Otari may collect and store agent session content to support session monitoring, troubleshooting, and visibility into agent activity. This collection is optional and depends on your workspace administrator's settings and your agent's configuration. When enabled, session content may include prompts, tool inputs and file paths, information evaluated by guardrails, and, if your workspace turns them on, what the agent's tools returned and the responses generated by the AI models your agent uses. Session content is encrypted, subject to access controls, and retained for a limited period, as described below. If an owner or administrator of your organization or workspace turns on session content, and your agent is configured to send it, Mozilla.ai will store the prompt that starts each turn of an agent session, the inputs the agent gives its tools (such as commands, file paths and the contents of files it writes or edits), and the information our guardrail checks evaluated. If your workspace also turns on storing tool results, Mozilla.ai will store what those tools returned. If your workspace also turns on storing model responses, Mozilla.ai will store the responses generated by the AI models during the session. Tool results and model responses are not stored unless these settings are on. Everything listed here is stored as session content: it is encrypted, subject to the same access controls and sharing, and deleted on the same schedule as the rest of the session You can choose to share session content with your organization's owners and administrators. Session content sharing ends 30 days after the session’s last activity, when the session is deleted (within 90 days of its start), or when you leave the workspace or organization, and does not resume if you return. If an owner of your organization enables session content, your organization's owners and administrators, and Mozilla.ai personnel can see that a session took place and what it cost, but the Service does not show them its content. Organization owners can open any session, and administrators can read content only through a share. Organization owners can access the content directly without submitting an access request. However, they must provide a reason for accessing it, which is recorded in the access logs for both owners and Mozilla.ai. **Administrative access to session content**. Your organization's owners can open the content of any of the organization's sessions regardless of whether you have shared the content, and Mozilla.ai can do so as well where required to comply with the law or legal process or where necessary to protect the rights or safety of others. Mozilla.ai records a reason for each such access. When your organization's owners access content, your session content may be visible to other owners/admins for your organization, but not to Mozilla.ai. Access to session content by Mozilla.ai is visible to Mozilla.ai's operators, but the content is not shown to your organization (unless accessed by an owner/admin, separately). Users are not notified when an owner/admin or Mozilla.ai has accessed unshared content. Records of such access are retained for one year (or longer where required by law or necessary to protect the rights, property, or safety of Mozilla.ai or others), including after your account or your organization is deleted. Session content is deleted together with the session 30 days after its last activity and 90 days after it started at the latest, when a workspace administrator deletes it, or when your user is deleted. Copies may persist in our encrypted backups until those backups expire. Agent session content (and associated provider credentials) are encrypted at rest. The keys are managed in AWS Key Management Service and can be used only by the running Service, and AWS records each time the Service obtains one. Access to session content is controlled by the Service and recorded; our operational procedures restrict and record direct access to the production environment. For more information about agent session content, please refer to our [Privacy Policy](https://otari.ai/privacy). --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.otari.ai/privacy/agent-session-content.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.