How do I use Elba MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Elba MCP Server list all my agents" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Elba MCP Server

by Kolsetu-Opensource

Overview Schema Related Servers Score Discussions

TypeScript

Remote

NOTE

Early preview — This project is under active development. See the roadmap below for planned tools.

Why?

Building voice AI agents usually means clicking through a dashboard. With this MCP server, you can do it all from your editor:

Create agents with a prompt — describe what you want, your AI assistant builds it
Iterate fast — change system prompts, swap voices, test — all without leaving your IDE
Version control — agent configs live in your workflow, not trapped in a dashboard
Zero context switching — ask Claude or Cursor to manage your agents directly

Elba is a voice AI agent platform by Kolsetu. This MCP server is a thin client that translates MCP tool calls into Elba API requests — all business logic lives in the Elba platform.

Related MCP server: mcacp

Quick Start

1. Get your integration token

Go to Elba Dashboard → Settings → Integrations → Copy your token.

2. Add to your MCP client

claude mcp add elba -- -e ELBA_INTEGRATION_TOKEN=your-token-here npx -y @kolsetu-opensource/elba-mcp-server

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "elba": {
      "command": "npx",
      "args": ["-y", "@kolsetu-opensource/elba-mcp-server"],
      "env": {
        "ELBA_INTEGRATION_TOKEN": "your-token-here"
      }
    }
  }
}

Add to .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "elba": {
      "command": "npx",
      "args": ["-y", "@kolsetu-opensource/elba-mcp-server"],
      "env": {
        "ELBA_INTEGRATION_TOKEN": "your-token-here"
      }
    }
  }
}

Add to .vscode/settings.json:

{
  "mcp": {
    "servers": {
      "elba": {
        "command": "npx",
        "args": ["-y", "@kolsetu-opensource/elba-mcp-server"],
        "env": {
          "ELBA_INTEGRATION_TOKEN": "your-token-here"
        }
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "elba": {
      "command": "npx",
      "args": ["-y", "@kolsetu-opensource/elba-mcp-server"],
      "env": {
        "ELBA_INTEGRATION_TOKEN": "your-token-here"
      }
    }
  }
}

3. Start using it

Once connected, ask your AI assistant things like:

"List all my Elba agents"
"Create a new voice agent called 'Support Bot' that handles customer billing questions"
"What voices are available? Switch my agent to use a different voice"
"Update the system prompt for my agent to be more friendly"

Available Tools

Tool	Description
`list_agents`	List all agents in your organization — returns ID, name, and status
`list_voices`	Browse available voices — returns ID, name, description, and provider
`upsert_agent`	Create or update a voice AI agent — matches by name or explicit ID

Architecture

┌─────────────────┐     stdio      ┌──────────────────┐     HTTPS     ┌──────────────┐
│   MCP Client    │◄──────────────►│  Elba MCP Server │◄────────────►│  Elba API    │
│ (Claude, Cursor │                │  (this project)  │              │  (backend)   │
│  VS Code, etc.) │                │  Thin client     │              │              │
└─────────────────┘                └──────────────────┘              └──────────────┘

Thin client — no business logic here; the server translates MCP calls into API requests
Local execution — runs as a stdio child process, no network exposure
Org-scoped tokens — each integration token is limited to a single organization

Roadmap

Phase 1 — Core Agent Management ✅

Tool	Description	Status
`list_agents`	List all agents in your organization	✅ Done
`upsert_agent`	Create or update a voice AI agent	✅ Done
`list_voices`	Browse available voices with descriptions	✅ Done

Phase 2 — Deploy & Test

Tool	Description	Status
`deploy_widget`	Deploy agent as embeddable web widget	Planned
`trigger_test_call`	Trigger a test call to your agent	Planned
`get_call_history`	View recent call history and outcomes	Planned
`get_credits`	Check organization credit balance	Planned

Phase 3 — IVR & Telephony

Tool	Description	Status
Workflow tools	IVR replacement, call routing	Future
Phone number management	Twilio BYOC import, number assignment	Future

Security

This project takes supply chain security seriously:

Measure	Status
OpenSSF Scorecard
CodeQL Analysis	Static analysis on every push & PR
OSV Scanner	Dependency vulnerability scanning via Google OSV
Dependency Review	Block PRs introducing known-vulnerable packages
StepSecurity Harden Runner	Audit all outbound network calls in CI
Pinned Actions	All GitHub Actions pinned to full SHA commits
npm Provenance	Published with npm provenance for verifiable builds
Signed Commits	DCO sign-off required on all contributions

Authentication

Log in to Elba
Go to Settings → Integrations
Generate or copy your integration token
Set it as ELBA_INTEGRATION_TOKEN in your MCP client config

Tokens are scoped to your organization — they can only access agents and data within your org.

Vulnerability Reporting

Do NOT open public issues for security vulnerabilities. See SECURITY.md for responsible disclosure via GitHub Security Advisories.

Development

# Clone the repo
git clone https://github.com/Kolsetu-Opensource/elba-mcp-server.git
cd elba-mcp-server

# Install dependencies
npm install

# Build
npm run build

# Run tests
npm test

# Type-check
npm run typecheck

# Lint
npm run lint

# Start in dev mode
npm run dev

Requires Node.js ≥ 20 (see .nvmrc).

Project Structure

src/
├── index.ts          # CLI entry point
├── server.ts         # MCP server init, tool registration
├── config.ts         # Environment config
├── auth.ts           # Token format validation
├── elba-client.ts    # Elba API client
└── tools/
    ├── list-agents.ts
    ├── list-voices.ts
    └── upsert-agent.ts
__tests__/            # Mirrors src/ structure (vitest)

Troubleshooting

"ELBA_INTEGRATION_TOKEN environment variable is required"

Make sure you've set the token in your MCP client config. See Quick Start for examples.

"Invalid or expired integration token"

Your token may have been revoked or expired. Generate a new one from Elba Dashboard → Settings → Integrations.

Server not starting

Check you're running Node.js ≥ 20: node --version
Try running directly: ELBA_INTEGRATION_TOKEN=your-token npx @kolsetu-opensource/elba-mcp-server
Check your MCP client logs for error messages

Still stuck?

Open a discussion for questions
File an issue for bugs

Contributing

Contributions are welcome! See CONTRIBUTING.md for guidelines.

Quick version:

# Make your changes, then:
npm run lint && npm run typecheck && npm test
npx changeset    # describe your change
git commit -s     # sign-off required (DCO)

License

MIT — see LICENSE for details.

Available Tools

3 tools

list_agentsA

List all voice AI agents in your Elba organization. Returns each agent's ID, name, and status (active/inactive).

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4.5/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the burden. It discloses that the tool lists agents and returns specific fields (ID, name, status). It does not mention any destructive behavior or side effects, which is acceptable for a read-only list operation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single sentence with two concise clauses. Every word adds value. It is front-loaded with the action and resource, and efficiently communicates the return information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with zero parameters and no output schema, the description is complete. It fully explains the tool's purpose and output, leaving no gaps for an AI agent to misunderstand.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema has zero parameters, and schema description coverage is 100%. Per guidelines, with 0 parameters the baseline is 4. The description adds no parameter information, which is appropriate since there are none.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the verb 'List', the resource 'voice AI agents', the scope 'in your Elba organization', and the returned fields (ID, name, status). This distinguishes it from sibling tools like 'list_voices' which would list voices, and 'upsert_agent' which is for create/update.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implicitly indicates when to use this tool (when you need a list of all agents). While it does not explicitly state when not to use it or name alternatives, the sibling tool names provide differentiation. For such a simple tool, the usage context is clear.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

list_voicesA

List available voices for Elba voice AI agents. Returns voice ID, name, description, and provider. Use the voice ID as the "voice" parameter in upsert_agent.

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4.4/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It correctly implies a read-only operation and specifies what fields are returned. It does not mention destructive behavior, which is appropriate for a list operation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two efficient sentences with no wasted words. The key information about purpose, return data, and usage guidance is front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given no parameters and no output schema, the description adequately explains the tool's output and connection to upsert_agent. It does not mention pagination or edge cases, but for a simple list tool, this is sufficient.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

There are zero parameters, and schema description coverage is 100%. The description adds meaning beyond the schema by detailing what is returned, meeting the baseline for a no-parameter tool.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states 'List available voices for Elba voice AI agents' with a specific verb and resource. It distinguishes from siblings list_agents and upsert_agent by focusing on voices.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly advises to use the returned voice ID as the 'voice' parameter in upsert_agent, providing clear context for when to use this tool. It does not state when not to use, but the sibling alternatives imply usage boundaries.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

upsert_agentA

Create or update a voice AI agent. If an agent with the given name exists, it will be updated. Otherwise, a new agent is created with sane defaults.

ParametersJSON Schema

Name	Required	Description
`name`	Yes	Agent name (used for upsert matching)
`voice`	No	Voice ID from list_voices (e.g., "alloy", "coral"). If omitted on create, defaults to "alloy". If omitted on update, preserves current voice.
`agent_id`	No	Explicit agent ID to update directly. If provided, skips name-based matching. Use list_agents to find agent IDs.
`system_prompt`	Yes	System prompt defining the agent's behavior

TDQS

A4.4/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, so description carries full burden. It discloses the upsert logic, name-based matching, and default voice behavior. It does not mention auth or error states, but for a simple upsert, this is adequate.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two succinct sentences that convey all essential information without unnecessary words. Perfectly front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given no output schema and no annotations, the description covers the upsert logic and parameter behavior sufficiently. It lacks mention of return value or error conditions, but these are minor gaps for a tool of this complexity.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, baseline 3. The description adds valuable context: name is for upsert matching, voice defaults differ on create vs update, and agent_id skips name matching. This goes beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states 'Create or update a voice AI agent', specifying the verb (create/update) and resource (agent). It distinguishes from siblings list_agents and list_voices, which are listing tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explains the upsert behavior: 'If an agent with the given name exists, it will be updated. Otherwise, a new agent is created with sane defaults.' This provides clear context for when to use, but lacks explicit when-not-to-use or alternative recommendations.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

3 tool updatesv0.1.0Jul 10, 2026
- First observedlist_agents
- First observedlist_voices
- First observedupsert_agent

Install Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.5/5.0

Disambiguation5/5

Each tool targets a distinct resource: list_agents for agents, list_voices for voices, upsert_agent for creating/updating agents. No functional overlap.

Naming Consistency5/5

All tools use a consistent verb_noun pattern (list_agents, list_voices, upsert_agent) with clear, predictable naming.

Tool Count5/5

Three tools is well-scoped for a voice AI agent management server, covering listing agents and voices, and creating/updating agents without unnecessary clutter.

Completeness4/5

The set covers the main lifecycle operations (list, create/update) but lacks a delete tool, which may be needed for full management. However, upsert covers creation and updates adequately.

Maintenance

ActivityInactive

ResponsivenessSyncing

Resources

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Chat ThingOAuth
ai.chatthing
Build and manage AI-native customer support agents from Claude or any MCP client.
Customer Support Autonomous Agents AI & Machine Learning
2
Apuchat A
io.github.opcastil11
Real-time chat for AI agents. Claude Code, Cursor, Cline and Codex join channels over MCP.
Communication Agent Orchestration
181
RogerRat A
io.github.opcastil11
Real-time chat hub for AI agents — Claude Code, Cursor, Cline, Codex over MCP or REST.
Communication Agent Orchestration
21
QuallaaOAuth
com.quallaa
Talk to your public-facing AI from any MCP client — Claude, ChatGPT, Cursor, Cline, Windsurf.
AI & Machine Learning Communication
2

Related MCP Servers

all-agents-mcp
Agent Orchestration Code Analysis Coding Agents
Dokkabei97
A
license
B
quality
F
maintenance
Enables orchestrating multiple AI CLI agents (Claude Code, Codex, Gemini CLI, Copilot CLI) through a unified MCP interface for task delegation, cross-agent comparison, and specialized tools like code review and debugging.
Updated 6 months ago (2026-03-15 05:49 UTC)
14
13
14
MIT
mcacp
Coding Agents Agent Orchestration Autonomous Agents
Oortonaut
A
license
A
quality
D
maintenance
Bridges any MCP client (like Claude Code, Zed, VS Code) to any ACP coding agent, enabling multi-agent orchestration from a single chat interface.
Updated 7 months ago (2026-02-01 05:08 UTC)
24
210
9
Apache 2.0
movoice-mcpofficial
Speech Processing AI & Machine Learning Communication
Movoiceai
A
license
Not graded
quality
C
maintenance
Create and manage voice AI agents using natural language from Claude Desktop or Cursor. Enables agent creation, listing, updating, deletion, and viewing call logs.
Updated 6 months ago (2026-03-19 19:39 UTC)
13
MIT
curatedmcp
Developer Tools Command Line Security
oneprofile-dev
A
license
A
quality
B
maintenance
A CLI agent that acts as an MCP hub, enabling discovery, auditing, and governance of MCP servers across AI tools like Claude and Cursor.
Updated a month ago (2026-07-29 18:24 UTC)
5
55
1
MIT

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Kolsetu-Opensource/elba-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

Why?

Quick Start

1. Get your integration token

2. Add to your MCP client

3. Start using it

Available Tools

Architecture

Roadmap

Phase 1 — Core Agent Management ✅

Phase 2 — Deploy & Test

Phase 3 — IVR & Telephony

Security

Authentication

Vulnerability Reporting

Development

Project Structure

Troubleshooting

"ELBA_INTEGRATION_TOKEN environment variable is required"

"Invalid or expired integration token"

Server not starting

Still stuck?

Contributing

License

Available Tools

TDQS

TDQS

TDQS

Tool Schema Changelog

TDQS

Maintenance

Resources

Looking for Admin?

Related MCP Connectors

Related MCP Servers

all-agents-mcp

mcacp

movoice-mcpofficial

curatedmcp

Latest Blog Posts

MCP directory API