Convai MCP for AI. Build living, knowledgeable NPCs for your project.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Convai lets you build entire conversational AI worlds from scratch. Use this MCP to create complex characters with specific backstories and voices for games or apps.
It manages narrative flow, stores deep knowledge in a bank, and allows your agent to test character responses in real-time.
What your AI can do
List chat sessions
Returns a list of previous conversations for a specific character.
List knowledge bank
Displays the current status and contents of your uploaded knowledge sources.
List languages
Shows all language codes that the character is capable of using.
Create new characters, update their details, or delete them entirely using specific actions.
Define plot points, create narrative sections, and set triggers that force the conversation down a specific storyline path.
Upload documents to the Knowledge Bank so your NPCs can answer complex questions based on specialized materials.
Review chat transcripts, list session details, and evaluate conversation quality against custom criteria.
Connect in real-time to test how a character responds using connect_live or generate sample conversations for testing.
Ask an AI about this
Waiting for input…
Convai: Character & Narrative Tools (25)
These tools let you manage the entire lifecycle of an AI character—from initial creation to live testing, story mapping, and knowledge grounding.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Convai on VinkiusList Chat Sessions
Returns a list of previous conversations for a specific character.
List Knowledge Bank
Displays the current status and contents of your uploaded knowledge sources.
List Languages
Shows all language codes that the character is capable of using.
List Narrative Sections
Retrieves a list of every narrative section you've created in your story map.
List Narrative Triggers
Lists all the triggers that control when and how a storyline changes.
List Voices
Returns an inventory of available character voices, categorized by provider.
Toggle Narrative
Turns the narrative-driven mode on or off for a specific character.
Connect Live
Connects your agent directly to the character's live API endpoint for real-time...
Create Character
Creates a brand new conversational AI character with specified traits and voices.
Create Narrative Section
Builds a structured section of the story that characters can follow or encounter.
Create Narrative Trigger
Sets up a specific event or condition that causes the narrative to shift or branch.
Delete Character
Removes an existing character profile from your account entirely.
Delete Knowledge Bank
Deletes a file or knowledge bank source you uploaded to the system.
Edit Narrative Section
Modifies the content and flow of an existing story section.
Evaluate Conversation
Runs a chat transcript through criteria to score or analyze its quality.
Generate Backstory
Generates detailed character descriptions and histories based on simple prompts.
Generate Starter Conversation
Creates multiple follow-up dialogue options to kick off a chat session or test an...
Get Character
Retrieves all specific details for one character by its ID.
Get Chat Session Details
Fetches the full record and history of a completed chat session.
Get Prompt
Retrieves configuration data for debugging or advanced setup checks.
Get Response
Forces the character to give a specific response based on an input prompt.
List Characters
Lists all characters that are currently active under your account.
Update Character
Makes changes to an existing character's traits, backstory, or voice settings.
Update Knowledge Bank
Replaces or modifies content within a previously uploaded knowledge source file.
Upload Knowledge Bank
Adds a new document or set of files to the character's permanent knowledge base.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Convai integration is available immediately — no restart needed.
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 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ 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
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 connection provides 25 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Writing believable characters manually is a nightmare of copy-paste.
Right now, building a world means keeping dozens of spreadsheets open: one for traits, one for backstories, another for key plot points. When you write the dialogue, you're constantly copying and pasting details from character sheets into your notes, hoping you don't accidentally contradict something established three chapters ago.
With this MCP, your agent handles that heavy lifting. Instead of managing spreadsheets, you just tell it to create a new character or map out a story trigger. You get characters whose traits are baked into their core logic; the complexity disappears right out of your workflow.
Control Character Lore with `generate_backstory`
Before, developing a character's history was an academic exercise: research, draft, revise. You spent days outlining their journey into separate documents that rarely spoke to each other.
Now, you use `generate_backstory`. You give the prompt—'a cynical space captain who loves jazz,' and instantly get rich, usable lore. It’s not just generating text; it's providing an actionable character framework for your project.
What your AI can actually do with this
This connection gives your agent full control over the Convai Engine. You can design entire digital beings; these aren't just simple chatbots. Instead, you build characters with defined backstories, unique voices, and deep expertise drawn from uploaded documents. If you're writing a massive RPG or building an app that needs believable NPCs, this is what you need.
You manage everything—from creating the initial character profile using generate_backstory to mapping out story triggers and sections for branching narratives. Need to ground your characters in real-world data? Upload documents into the Knowledge Bank and let them reference that information naturally during conversation. When all of this gets complicated, Vinkius makes it easy; you connect once from any MCP-compatible client and get access to the entire Convai toolset, letting your agent do the heavy lifting.
019e5d0b-62bc-72d4-b6f2-24f986262e07 
Here's how it actually works
The bottom line is you tell your AI agent what kind of characters and stories you want, and it handles the creation, storage, and deployment through structured calls.
Subscribe to the Convai MCP and provide your API key.
Connect your agent via any compatible client, granting it access to the character tools.
Use commands like create_character or upload_knowledge_bank to build out your world's lore and knowledge.
Who is this actually for?
This MCP serves game developers who can't rely on simple dialogue trees, professional writers building complex worlds, or researchers needing to test conversational agents against specific knowledge bases.
Needs to iterate quickly on NPC personalities and backstories directly from their development environment.
Must map out complex story triggers, sections, and plot points without leaving their writing workspace or jumping between multiple platforms.
Needs to evaluate conversation quality across many test scripts and manage knowledge banks for specialized conversational agents.
What Changes When You Connect
Develop complex lore instantly. Instead of spending hours drafting character profiles in a separate document, you can use generate_backstory to create detailed histories right from your agent's prompt window.
Manage story branching without pain. You define narrative sections and triggers (using create_narrative_section and create_narrative_trigger), telling the AI exactly when and how the plot must change, making your world feel cohesive.
Ground characters in specific facts. By uploading documents using upload_knowledge_bank, you ensure that conversations aren't just creative; they are factual, forcing the character to reference your source material.
Test interactions on demand. Need to check how a character responds before a live demo? Use generate_starter_conversation or connect via connect_live to test dialogue flow right away.
Audit and refine performance. After running beta tests, use evaluate_conversation to score transcripts against specific rules. This lets you know exactly which parts of the character's personality are inconsistent.
See it in action
Building a fantasy RPG companion
A game developer needs an NPC, 'Kael,' who must act like a grizzled mercenary. They use create_character to build him and then run generate_backstory, feeding the resulting lore back into the character's knowledge base via upload_knowledge_bank. Finally, they test his dialogue flow using get_response before handing it off to art.
Creating a complex historical simulation
A writer needs characters who only talk about 18th-century trade routes. They use the knowledge bank tools (upload_knowledge_bank) to feed in primary source documents and then set up narrative sections that force conversations around those specific topics.
Debugging customer service dialogue
A company researcher wants to ensure their AI support character never gives outdated advice. They list_knowledge_bank contents, update the necessary documents using update_knowledge_bank, and then use evaluate_conversation on old transcripts to check for compliance.
The honest tradeoffs
Treating characters as simple chatbots
Thinking that just writing a detailed description of the character is enough. If you don't define how they react when challenged, the conversation will feel random and unscripted.
You must use create_character to set the core voice, then build structure using create_narrative_trigger and list_narrative_triggers. This ensures their reactions are predictable and tied to the story's logic.
Forgetting context or facts
The character gives a cool answer, but it contradicts something established in the world lore. The conversation instantly loses credibility.
Never rely on memory alone. Always upload core documents using upload_knowledge_bank. This forces the agent to ground its answers in your specific source material.
Making changes without testing
You update a character's backstory (update_character) and assume it works, but you never see how that change affects their dialogue.
Always test the changes. Use get_response to get immediate feedback on how the updated lore impacts the actual output, or use connect_live for full flow testing.
When It Fits, When It Doesn't
Use this MCP when your project requires deep continuity and structured interaction. If you need an NPC that behaves like a person—one who remembers facts (upload_knowledge_bank), follows rules (narrative sections), and has distinct personality quirks (create_character)—this is the tool for you. However, if all you need is a simple Q&A bot using only general internet knowledge, this MCP might be overkill; a standard RAG service without narrative control would suffice. Don't use it just because it has many tools; use it when story structure and character depth are your primary goals.
Questions you might have
How do I start a new character with the Convai MCP using `create_character`? +
You call create_character and provide the initial details, like their name, voice, and core purpose. This establishes their profile in the system so you can begin building out lore around them.
I need to make sure my character uses specific company rules; which tool should I use? +
You must upload those rules or documents using upload_knowledge_bank. This ensures the character's knowledge base is restricted and accurate, forcing it to adhere to your source material.
What is the difference between `list_chat_sessions` and `get_chat_session_details`? +
list_chat_sessions just gives you a list of previous chats for a character. To see what actually happened in one specific chat, you must use get_chat_session_details.
How do I make sure my story has multiple possible endings? +
You define the flow using narrative tools. Create a main path with create_narrative_section, and then set branching points with create_narrative_trigger to guide the agent to different outcomes.
Can I test my character's voice live before finalizing it? +
Yes. You can connect directly using connect_live. This lets you run a real-time chat session and hear how the character's response sounds, which is key for quality control.
If I need to purge old training materials entirely, how do I use the `delete_knowledge_bank` tool? +
It permanently removes the specified knowledge bank file from your account. Always run list_knowledge_bank first to confirm the ID and status of the material you're removing.
What happens to linked assets, like narratives, when I use the `delete_character` tool? +
The character record itself is removed. However, any narrative sections or triggers associated with that ID remain available for re-assignment and reuse.
If I need specific technical metadata about a character, should I use the `get_character` tool? +
Yes, this fetches all detailed parameters for an existing character by its ID. This is crucial for debugging or passing precise data structures to other systems.
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.
We've already built the connector for Convai. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 25 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.