Cody AI MCP. Manage Knowledge-Base Bots Via API
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Cody AI. Manage your knowledge-base bots and document training via API. Your AI agent uses this to list, configure, and query custom bots trained on proprietary documents, making complex internal knowledge instantly available to conversation models.
What your AI agents can do
Create conversation
Starts a new conversation session with a specific bot.
Get bot details
Retrieves detailed information about a specific bot's configuration and training parameters.
Get document status
Checks if the AI has finished processing a document, confirming it's ready to answer questions.
The agent initiates a chat session with a specific bot, allowing the user to immediately ask questions against the bot's knowledge.
The agent retrieves the full configuration and status information for any specified bot.
The agent verifies if the AI has finished processing and learning the content of an imported document.
The agent takes a web address and sends the content to be added to a designated folder in the knowledge base.
The agent fetches a list of all active bots configured in the account, or retrieves details about a specific bot.
The agent lists folders, retrieves all documents, or gets the message history for any specific conversation.
The agent sends a specific prompt or question to an active bot within a designated conversation thread.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
019d7576create conversation
Starts a new conversation session with a specific bot.
019d7576get bot details
Retrieves detailed information about a specific bot's configuration and training parameters.
019d7576get document status
Checks if the AI has finished processing a document, confirming it's ready to answer questions.
019d7576import webpage
Sends content from a URL to be added to a specific folder in the knowledge base for bot training.
019d7576list bots
Fetches a list of all configured bots in your Cody AI account.
019d7576list conversations
Retrieves a list of recent conversation sessions for review.
019d7576list documents
Gets a list of all documents currently stored in the knowledge base.
019d7576list folders
Retrieves a list of folder containers within your knowledge base structure.
019d7576list messages
Retrieves the message history for a specific, existing conversation thread.
019d7576send message
Sends a prompt to a bot, generating a response using its trained knowledge.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Cody AI, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
You're gonna connect your AI agent to Cody AI. This server gives your agent tools to manage all your knowledge-base bots and train 'em on your own docs. Your agent can list all configured bots, check details on any specific bot, and start a conversation with one of them. You can also pull a list of all active conversation sessions, and check the message history for any thread you're reviewing.
Need to manage the content? Your agent can list all the folders in your knowledge base, get a list of every document stored, or even list all the messages in a specific conversation. Wanna feed the bot new info? Your agent can take a web address and send that content to a designated folder for bot training.
You can check if the AI finished processing a document using get_document_status before you use it. To actually use the bot, your agent can send a prompt to a bot, making it generate a response using its trained knowledge. You can also kick off a brand new conversation session with a specific bot, letting you ask questions against the bot's knowledge right away.
You can manage your bots by running list_bots to fetch all active bots in your Cody AI account, or use get_bot_details to pull the full setup and status for a specific one. Your agent handles all the heavy lifting so you don't gotta manually jump into the dashboard. It's built for anyone who needs to manage complex internal knowledge, whether you're a support team needing instant answers, a knowledge manager updating content, or a developer integrating these bots into a bigger workflow.
How Cody AI MCP Works
- 1 Add the Cody AI integration to your AI toolset and provide the API key.
- 2 Your agent calls a tool (e.g.,
list_bots) to read available data. - 3 The agent processes the response and can then call a second tool (e.g.,
create_conversation) to act on that data.
The bottom line is that your agent uses these tools to manage and query specialized, custom AI assistants trained on your private content.
Who Is Cody AI MCP For?
Knowledge Managers who need a way to track and ingest new corporate documents. Support Engineers who need instant, accurate answers to customer questions without manual searches. Developers building internal tools that need to reference proprietary documentation.
Uses the agent to query trained bots for immediate answers to customer issues, citing sources from the knowledge base.
Uses the agent to import new web content or files, and then manages the bot's training status via get_document_status.
Uses the agent to programmatically list bots (list_bots) and automate the creation of new conversation sessions (create_conversation) within workflows.
What Changes When You Connect
- Stop manual knowledge sharing. Use
import_webpageto feed entire web pages into a bot's knowledge base with a single call. - Gain full oversight of your AI assets.
list_botslets you see every active bot and its current training status at a glance. - Know when content is ready.
get_document_statusconfirms if the AI has finished indexing a file, preventing bad answers. - Maintain conversation history. Use
list_messagesto pull the full transcript of any session, perfect for auditing or handoffs. - Build reliable workflows. Combine
list_botsandcreate_conversationto automate the setup of new support channels. - Targeted querying.
send_messagedirects your AI agent to ask a specific bot a question, keeping the context clean and focused.
Real-World Use Cases
Onboarding a New Policy
A Knowledge Manager needs to update the Legal Bot. Instead of manually uploading PDFs, they use import_webpage to pull the new policy guidelines from a URL. The agent then monitors get_document_status until the bot is ready to answer questions.
Handling Complex Customer Inquiries
A Support Agent receives a complex query. The agent first uses list_bots to identify the 'Product FAQ' bot, then calls create_conversation, and finally uses send_message to get a source-cited answer for the customer.
Auditing Bot Performance
A Platform Developer needs to check if a bot is working correctly. They use list_messages to review the last 50 interactions, ensuring the bot is adhering to the expected policy and conversation flow.
Structuring Knowledge Bases
A Content Architect wants to organize documents. They use list_folders to see existing categories, and then use list_documents to confirm all required files are present before training a new bot.
The Tradeoffs
Assuming all bots are active
Calling send_message without first checking if the bot exists. The agent fails and the user gets a vague error about missing credentials.
→
Always call list_bots first to validate the bot's existence. Then, use get_bot_details to confirm its current status before initiating a create_conversation.
Ignoring document processing time
Calling send_message immediately after an upload. The AI hasn't indexed the new content, so the answer is generic and wrong.
→
Use import_webpage to add the content, but always check get_document_status first. Wait until the status confirms the document is fully learned.
Manually listing all contents
Trying to find a specific document's location by calling list_documents, then list_folders, and then cross-referencing the results manually in the chat.
→
Use the specific tools. Start with list_folders to narrow down the scope, then use list_documents to get the exact list, making the data structured for the agent.
When It Fits, When It Doesn't
Use this MCP Server if your core business problem involves making proprietary, documented knowledge available to an AI agent. You need to manage the lifecycle of specialized bots and ensure the data they use is fresh and vetted. You should use it when you need to: 1) ingest external web content (import_webpage), 2) check the status of data (get_document_status), or 3) programmatically manage conversations (create_conversation).
Don't use this if you just need general, public knowledge (use a general search API). Don't use this if you only need to send a single, isolated prompt without needing conversation context (use a simple text generation API). You need the entire workflow—the management and the querying. When in doubt, check list_bots first.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Cody AI. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Finding out what your bots know shouldn't require logging into three different dashboards.
Today, if you want to know if a bot is trained on the latest policy, you have to jump into the Cody AI dashboard. You check the 'Document Status' tab, copy the document ID, then switch to the 'Bot Configuration' tab to confirm which bot uses it. Finally, you check the 'Conversation Logs' to see if it's actively being used. It's a mess of tabs and copy-pasting.
With this MCP server, your agent handles it all. You ask, 'Is the Legal Bot ready for the Q3 policy update?' and the agent calls `get_document_status` and `get_bot_details` in one go. You get a single, clean answer telling you exactly what's ready and what's missing.
Cody AI MCP Server: Automate Bot & Document Ops
Manual document ingestion is slow. You copy the URL, paste it into a form, and wait for a progress bar to fill up. Then, you wait for the AI to finish vectorizing the data. This is a major bottleneck for fast support teams.
Now, the agent handles the ingestion. You just tell it to `import_webpage` the URL. The agent sends the content, and you can monitor the process using `get_document_status`—all from your natural language interface. It's instant control.
Common Questions About Cody AI MCP
How do I list all the bots configured using the `list_bots` tool? +
The list_bots tool retrieves a full list of all bots in your account. The response includes key metrics like the bot's name, document count, and conversation volume.
What is the difference between `list_documents` and `list_folders`? +
Use list_folders first to get the high-level structure of your knowledge base. Then, use list_documents to get the actual file names and metadata within those folders.
Can I ask a question using `send_message` if the bot is not trained? +
The system will tell you if the bot is trained on insufficient data. You must use get_document_status and import_webpage to add content before send_message works correctly.
How do I start a new chat with a specific bot using `create_conversation`? +
You call create_conversation and specify the target bot ID. This generates a unique session ID, which you then use for all subsequent send_message calls.
How do I check the history of a conversation? Use `list_messages`. +
The list_messages tool requires the specific conversation ID. It returns the full message history, allowing you to review who said what and when.
How do I check if a document is ready to be used with `get_document_status`? +
The get_document_status tool reports the syncing status of a document. If the status shows 'Complete' or 'Ready', the AI has finished processing and learning from the content.
What do I use `list_messages` for if I want to see the conversation history? +
You use list_messages to retrieve the message history for a specific conversation ID. This gives you the full context of what was discussed.
How do I ensure the knowledge base is up to date before I use `send_message`? +
To ensure the knowledge base is current, use import_webpage to add new URLs or list_documents to verify existing content. Always check the document status first.
How do I get a Cody AI API key? +
In the Cody AI dashboard, go to Settings > API and copy your key. Paste it below.
What is Focus Mode? +
Focus Mode restricts a bot's knowledge to specific documents for a given conversation, improving answer relevance.
How long does it take to learn new documents? +
Usually seconds to minutes depending on document size. Use 'get_document_status' to check when sync is complete.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
PagerDuty Incident Trigger
This MCP does exactly one thing: it triggers incidents in PagerDuty. That's its only function, and nothing else. Incredible for giving your AI agents the power to wake up engineers.
Delighted
Equip your AI agent to monitor customer feedback, track NPS metrics, and manage survey responses via the Delighted API.
WooCommerce Order Status Reader
This MCP gives your AI agent the ability to check the real-time status and tracking information of any WooCommerce order using an Order ID or a Customer Email. Perfect for automating "where is my order?" (WISMO) support queries.
You might also like
Avalara (AvaTax)
Automate tax calculation and compliance with Avalara AvaTax — calculate rates and resolve addresses via AI.
GetResponse
Manage email marketing campaigns, track subscribers, and oversee newsletters via AI agents with GetResponse.
Datadog
Monitor applications via Datadog — query performance metrics, search logs, and list active monitors directly from any AI agent.