mirror of
https://github.com/tiennm99/DocsGPT.git
synced 2026-10-14 18:18:08 +00:00
Document connectors and the upgrade to encrypted connections
- Guides > Connectors: using connections, sharing modes, attribution, and admin setup (encryption key and rotation, redirect URIs, Google Drive, SharePoint, Confluence, S3, MCP presets). - Integration pages register CONNECTOR_REDIRECT_BASE_URI as-is; the ?provider= suffix never matched what the backend sends. VITE_GOOGLE_CLIENT_ID is optional. - Upgrading: set ENCRYPTION_SECRET_KEY before migrating multi-user installs.
This commit is contained in:
1 parent
d9f823d601
commit
8f00118afc
9 files changed
+196
-19
No files matched your search
@@ -121,7 +121,8 @@ Interacting with tools in DocsGPT is designed to be intuitive:
|
||||
|
||||
2. **Configuration in UI:**
|
||||
* Tools are generally managed and configured within the DocsGPT application's settings, found under a "Tools" section in the GUI.
|
||||
* For tools that interact with external services (like Brave Search, Telegram, or any service via the API Tool), you might need to provide authentication credentials (e.g., API keys, tokens) or specific endpoint information during the tool's setup in the UI.
|
||||
* Tools for external services (Brave Search, Telegram, ntfy, PostgreSQL, MCP servers) are set up from **Settings > Connectors**. You connect the service once; its credentials stay encrypted on the server and never reach the browser, and the tools it provides are grouped under that connection. Each action is marked **Read** or **Write** and can be set to **Always allow**, **Needs approval** or **Off**. See [Connectors](/Guides/Connectors).
|
||||
* When a tool's connection is missing or expired, the chat shows a **Connect** prompt and continues once you connect.
|
||||
|
||||
3. **Prompt Engineering for Tools:** While the LLM aims to intelligently use tools, for more complex or reliable agent-like behaviors, you might need to customize the system prompts. Modifying the prompt can guide the LLM on when and how to prioritize or chain tools to achieve specific outcomes, especially if you're building an agent designed to perform a certain sequence of actions every time. For more on this, see [Customising Prompts](/Guides/Customising-prompts).
|
||||
|
||||
|
||||
Reference in new issue
Block a user