What is Confluence?
Confluence is a docs & notes tool. With OSP.net, your AI agent connects to it directly so it can act on your behalf instead of just answering questions about it.
What your OSP agent can do with Confluence
- Search and read pages and spaces
- Summarize space content
- Pull page bodies for drafting
How to connect Confluence
- Create an Atlassian API token (works for both Jira and Confluence): Atlassian account → Security → API tokens → Create API token. Copy it (shown once). id.atlassian.com → Security → API tokens
- Your Confluence URL is your site with /wiki on the end — e.g. https://your-site.atlassian.net/wiki.
- Paste your URL, your Atlassian account email and the API token here, then Connect. Your agent runs the Atlassian MCP server in READ-ONLY mode (no edits). Atlassian MCP server (read-only)
Security: Defense in depth: the agent always runs the Atlassian server with READ_ONLY_MODE on, so it can search and read but never edit pages. The API token still acts as your account on the spaces it can see — revoke it any time from the same Atlassian page.
Who uses Confluence with OSP
These roles commonly connect Confluence to their OSP agent:
Related docs & notes integrations
Frequently asked questions
- Can OSP.net connect to Confluence?
- Yes. Confluence is a native OSP.net integration — you bring your own Confluence key or token, paste it in your dashboard, and your agent restarts live.
- What can my OSP agent do with Confluence?
- Your agent reads and searches your Confluence Cloud spaces — read-only — via the official-grade Atlassian MCP server. Specifically: Search and read pages and spaces; Summarize space content; Pull page bodies for drafting.
- Is Confluence a native integration or via the gateway?
- Confluence is a native, baked-in integration. You connect it with your own credentials, which are stored in an encrypted vault and injected only at runtime.
- Is my Confluence data secure with OSP.net?
- Yes. Your Confluence credentials live in an encrypted secrets vault, are injected only at container runtime, and are never written to disk in plaintext or used to train any model. Each customer runs in a fully isolated instance.