Convai MCP. Build and manage entire character worlds, not just single lines of dialogue.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Convai MCP Server connects your AI client to the Convai Engine. You can build and manage complex, lifelike conversational AI characters (NPCs) for games and apps.
Use it to create backstories, map branching story triggers, upload knowledge documents, and test character dialogue in real time. It's your central hub for designing digital beings and their entire narrative world.
What your AI agents can do
Connect live
Establishes a real-time connection to the character for immediate voice testing via WebRTC.
Create character
Creates a brand new character profile in the system.
Create narrative section
Adds a new, defined segment to the overall story structure.
Creates, retrieves, updates, or deletes entire character profiles, including voice and backstory information.
Defines interconnected narrative sections and the specific triggers that move the story from one point to the next.
Uploads and manages external documents, giving characters deep, factual expertise for accurate responses.
Generates character responses, starter conversations, and connects live to test character performance and voice quality.
Retrieves and evaluates transcripts, allowing you to measure conversation quality against specific metrics.
Lists and fetches all existing characters, narrative sections, and knowledge bank statuses.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
019e5d0bconnect live
Establishes a real-time connection to the character for immediate voice testing via WebRTC.
019e5d0bcreate character
Creates a brand new character profile in the system.
019e5d0bcreate narrative section
Adds a new, defined segment to the overall story structure.
019e5d0bcreate narrative trigger
Establishes a condition that moves the story to a new narrative section.
019e5d0bdelete character
Permanently removes an existing character profile.
019e5d0bdelete knowledge bank
Removes a specific knowledge bank file from the system's memory.
019e5d0bedit narrative section
Modifies the content or parameters of an existing story segment.
019e5d0bevaluate conversation
Runs a transcript against custom rules to measure and grade the quality of a conversation.
019e5d0bgenerate backstory
Creates a detailed character description or background story based on a prompt.
019e5d0bgenerate starter conversation
Generates initial dialogue options to kick off a chat session with a character.
019e5d0bget character
Retrieves all current details for a specific, existing character ID.
019e5d0bget chat session details
Fetches the full transcript and metadata for a specific chat session.
019e5d0bget prompt
Retrieves the underlying prompt data for debugging or advanced configuration checks.
019e5d0bget response
Requests a specific reply from a character based on the current context.
019e5d0blist characters
Returns a list containing the IDs and names of all characters managed by the account.
019e5d0blist chat sessions
Retrieves a list of all past chat sessions for a specific character.
019e5d0blist knowledge bank
Shows the current status and files uploaded to the knowledge bank.
019e5d0blist languages
Returns a list of all languages the system supports for character dialogue.
019e5d0blist narrative sections
Lists all existing story sections defined in the system.
019e5d0blist narrative triggers
Lists all active story triggers that govern narrative movement.
019e5d0blist voices
Returns a categorized list of available synthetic voices from various providers.
019e5d0btoggle narrative
Switches a character between standard dialogue mode and narrative-driven story mode.
019e5d0bupdate character
Modifies any aspect of a previously created character, such as their backstory or voice.
019e5d0bupdate knowledge bank
Replaces or changes the content of an existing knowledge bank file.
019e5d0bupload knowledge bank
Adds a new document to the character's knowledge base for reference.
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 Convai, 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
Yo, this Convai MCP Server connects your AI client straight to the Convai Engine. It's your central hub for building and running complex, lifelike characters—the kind of NPCs you actually want in your games and apps. You'll use it to build backstories, map out branching story triggers, upload documents, and test character dialogue in real time.
It’s the whole damn setup for designing digital beings and their narrative world.
Managing Character Details
create_character lets you build a brand new character profile. You can then update_character to tweak anything about them, like their voice or backstory. Need to wipe them out? delete_character takes them permanently. To check on any character, use get_character for all their current details, or list_characters to see a list of every character ID and name you've got.
Building Story Arcs and Triggers
Want a story that actually flows? You use create_narrative_section to add a new, defined segment to the story structure. Then, you map out the movement with create_narrative_trigger, which sets a condition that moves the story to the next section. You can modify existing story bits with edit_narrative_section and flip the character between standard dialogue mode and story mode using toggle_narrative.
You can also list all story sections with list_narrative_sections and all active triggers with list_narrative_triggers.
Injecting Domain Knowledge
Don't let your characters sound dumb. You upload_knowledge_bank to add new documents, giving them deep, factual expertise. If you need to swap out the docs, update_knowledge_bank changes the content of an existing bank. You can check what's loaded with list_knowledge_bank or see all the languages the system supports for dialogue using list_languages}.
Simulating and Testing Dialogue
Ready to test the chat? You generate_starter_conversation to kick off a session, or get_response to get a specific reply based on the current context. You can run a live test using connect_live to immediately test the character's voice via WebRTC. To see what's happened before, list_chat_sessions gives you a list of past chats, and get_chat_session_details fetches the full transcript and metadata.
You can also generate_backstory to create a detailed character background or get_prompt to retrieve the underlying prompt data for debugging.
Analyzing Conversation Quality
You don't just guess if the dialogue slaps. You use evaluate_conversation to run a transcript against custom rules, measuring and grading the conversation quality. You can also fetch the current status and files in the knowledge bank with list_knowledge_bank and see all available synthetic voices from various providers by running list_voices.
How Convai MCP Works
- 1 1. Subscribe to the Convai MCP Server and input your API Key.
- 2 2. Your AI client calls the desired tool (e.g.,
create_character) to initiate a task. - 3 3. The server executes the action, modifying the character, narrative, or knowledge bank, and returns the result to your AI client.
The bottom line is, you tell your AI client what you want—a new character, a story section, or a response—and it handles the API calls to make it happen.
Who Is Convai MCP For?
Game developers and narrative designers need this. If you spend time writing lore or creating complex character interactions, this is for you. It lets you build and test entire worlds, not just single lines of dialogue. It’s for people who need system-level control over digital beings.
Uses create_character and update_character to define NPC personalities and backstories, ensuring they stay consistent across different game scenes.
Maps out complex story triggers using create_narrative_trigger and structures the overall story flow by managing sections with create_narrative_section.
Tests conversation quality using evaluate_conversation and manages specialized data sets via upload_knowledge_bank to refine agent behavior.
What Changes When You Connect
- Create and iterate on NPC personalities instantly. Use
create_characterandupdate_characterto define backstories, voices, and traits, keeping your characters consistent from the first line to the last. - Map complex story logic without leaving your editor.
create_narrative_sectionandcreate_narrative_triggerlet you build branching, non-linear storylines that react to user input. - Give characters real-world expertise.
upload_knowledge_banklets you feed the character proprietary documents, ensuring every response is grounded in your specific domain knowledge. - Test dialogue in a live environment. Use
get_responseorgenerate_starter_conversationto test character replies and voice fidelity before committing them to the final build. - Track and improve performance. After a playtest, use
list_chat_sessionsandevaluate_conversationto pull up transcripts and grade the conversation quality against specific metrics. - Maintain a single source of truth. Tools like
get_characterandlist_charactersgive you an immediate overview of every asset in your character library.
Real-World Use Cases
The RPG World Builder
A game developer needs a mysterious vendor character. Instead of writing static dialogue trees, they use the agent to create_character, give the vendor lore via upload_knowledge_bank, and define the vendor's interaction rules using create_narrative_section. They then use connect_live to test the character's voice and dialogue flow in real time.
The Educational Simulation
An AI researcher needs a historical figure character. They use create_character and list_languages to ensure the character speaks the correct dialect. They upload academic papers via upload_knowledge_bank and then run evaluate_conversation to confirm the character's responses are accurate to the source material.
The Interactive Novelist
A writer is stuck on a plot point. They use the agent to generate_backstory for a new character, then use create_narrative_trigger to map out three potential outcomes. They use list_narrative_sections to visualize the whole branching story map before writing a single word.
The Tech Support Bot
An operations team uses the agent to define a specialized bot persona via create_character. They feed the bot all internal documentation via upload_knowledge_bank and use get_response to ensure the bot answers specific, tricky questions accurately.
The Tradeoffs
Thinking of characters as static profiles
Just calling get_character to read the backstory, then leaving it there. This only gives you data; it doesn't let you interact with the character or test their dialogue.
→
Always test the character. Use generate_starter_conversation to get sample dialogue, or get_response to force a reply. This confirms the character's behavior, not just its profile data.
Ignoring the story flow
Writing a huge chunk of lore and uploading it via upload_knowledge_bank, but forgetting to link it to any plot point. The knowledge sits there, but the character never encounters it.
→
Use create_narrative_trigger to ensure the knowledge is relevant. Map the knowledge to a specific create_narrative_section so the story must pass through that moment to use the data.
Debugging manually
Trying to figure out why a character responded weirdly by guessing which API call failed. This is slow and non-deterministic.
→
Check the audit trail. Use list_chat_sessions to see the full history, and get_chat_session_details to review the exact context and conversation flow that led to the unexpected reply.
When It Fits, When It Doesn't
Use this server if your project requires the character to be more than just a name and a description. You need the character to act consistently and remember complex facts. Use it to build digital beings with systemic memory and narrative depth.
Don't use it if you just need a simple database entry for character names or simple, fixed dialogue trees. If your needs are limited to a single, predictable conversation flow, a basic database or simple message tool might suffice. But if you need to manage the interplay between lore, action, and knowledge—that's where Convai shines. Always check list_languages first if you're working internationally.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Convai. 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 25 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Building believable characters shouldn't feel like managing a dozen separate databases.
Right now, if you build an NPC, you have to manage a character sheet in one tool, a dialogue flow chart in another, and a separate document database for their lore. Then, if they say something that contradicts a document you uploaded, you have to manually check three different systems to find the error.
With the Convai MCP Server, you define the character using `create_character`. You manage the entire story structure using `create_narrative_section` and link all the necessary facts via `upload_knowledge_bank`. Your AI client handles the complex cross-referencing, giving you one single, coherent source of truth.
Convai MCP Server: Get live, multi-stage character responses.
Manually testing dialogue is a nightmare. You write a prompt, run it, get a response, then manually adjust the character's backstory, and run it all again. This iteration cycle is slow and tedious.
Now, you can use `get_response` to test a specific reply, or connect live with `connect_live` to run a full, real-time conversation. The system manages the state changes and context shifts automatically, letting you focus purely on the story.
Common Questions About Convai MCP
How do I list all my characters using the `list_characters` tool? +
Run list_characters to get a list of all IDs and names. This is the first step before you can fetch details or update any character.
What is the difference between `get_response` and `get_chat_session_details`? +
get_response gets a single, immediate reply from the character based on the current context. get_chat_session_details pulls the full, historical transcript of a conversation.
Do I need to use `upload_knowledge_bank` before I can use `create_character`? +
No, you don't. You can create the character profile first. But if you want the character to know specific facts, you must use upload_knowledge_bank afterwards.
How do I check if a character's voice is supported? +
Use list_voices to see all available voices. This tool returns them categorized by their provider, so you know exactly what's available.
When should I use `toggle_narrative` to control a character's story mode? +
You use toggle_narrative when you need to switch a character between standard conversation and a structured, story-driven mode. This lets you activate or deactivate the complex narrative rules that govern their dialogue flow.
How do I update a character's details using `update_character`? +
You call update_character by providing the character ID and the new data points you want to change. This allows you to modify anything from their core backstory to their voice settings without deleting and recreating the whole profile.
If I upload new data, how do I keep my knowledge up to date with `update_knowledge_bank`? +
Use update_knowledge_bank to revise existing files or add corrections to a document already in the bank. This ensures your characters pull the most current and accurate information when generating responses.
What information can I get about my conversation history using `list_chat_sessions`? +
The list_chat_sessions tool returns a summary list of all past interactions for a given character. You can then use get_chat_session_details to pull the full transcript and metadata for any specific session.
How do I create a new character with a specific voice? +
Use the create_character tool. You can specify the charName, backstory, and voiceType. To see available voices, run the list_voices query first.
Can I give my character custom documents to learn from? +
Yes! Use the upload_knowledge_bank tool to add files. Once uploaded, you can link them to your character using the update_character tool by passing the document IDs in the docs field.
How can I test a character's response to a specific input? +
You can use the get_response tool. Provide the charID and the user text, and the agent will return the character's generated response along with any triggered actions.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Date Utils Engine
Stop AI from hallucinating dates, missing leap years, or failing timezone conversions. Offloads all calendar math to the deterministic date-fns engine.
NVIDIA API Catalog
Cloud Engine proxy running native foundational completions natively utilizing active Nemotron and Llama3 architectures.
Neptune.ai (ML Experiment Tracking)
Manage ML experiments via Neptune.ai — track training runs, monitor metrics, and audit model versions.
You might also like
SWAPI - The Star Wars
Explore the Star Wars universe — query people, planets, films, and starships directly from the Star Wars API.
Wati
Send WhatsApp messages at scale with official API access, chatbot automation, and shared team inbox for business conversations.
idloom
Manage events, attendees, and invoices via idloom.events API.