Skip to main content
Glama

Eigi.ai Voice Agents

Server Details

Create and manage AI voice agents, real-time conversations, and analytics with eigi.ai

Status
Healthy
Last Tested
Transport
Streamable HTTP
URL

Available Tools

23 tools
create-agentTry in Inspector

Create a new voice agent. IMPORTANT: Before calling this tool, you MUST call 'list-providers' to get valid provider/model combinations for STT, LLM, and TTS, and 'list-voices' to get valid voice_id values for the chosen TTS provider. Do not guess provider names, model names, or voice IDs.

ParametersJSON Schema
NameRequiredDescriptionDefault
llmYesLLM config: provider_name, model_name, params.temperature
sttYesSTT config: provider_name, model_name, language
ttsYesTTS config: provider_name, model_name, language, voice_id
toolsNoConfigure agent tools: end_call, transfer_call, schedule_call, set_language, background_audio, user_idle
promptNoReference existing prompt by name instead of prompt_content
agent_nameYesAgent name, 1-100 characters
agent_tipsNoUsage tips array
agent_typeNoDefault OUTBOUND. Only set INBOUND if user explicitly requests it.OUTBOUND
widget_configNoWidget configuration for chat interface customization
agent_categoryNoCategory, default Other
prompt_contentNoSystem prompt text, creates internal prompt
agent_descriptionNoDescription, max 500 characters
agent_mobile_numberNoMobile number for the agent as a single-element array, e.g. ["918035315555"]. Number format: country code + number, no '+' prefix. IMPORTANT: Call 'get-mobile-numbers' first to get valid purchased numbers. INBOUND agents require unique numbers not assigned to other INBOUND agents.
first_message_promptNoGreeting message when call/chat starts
create-promptTry in Inspector

Creates a new prompt. If prompt name exists, creates a new version with auto-incremented version number.

ParametersJSON Schema
NameRequiredDescriptionDefault
prompt_nameYesPrompt name, 1-100 chars, e.g., 'customer_support'
prompt_typeNo'main' for agent prompts, 'analysis' for analysismain
prompt_contentYesSystem prompt text content
delete-agentTry in Inspector

Permanently delete an agent and its internal prompt. Cannot be undone.

ParametersJSON Schema
NameRequiredDescriptionDefault
agent_idYesAgent UUID to delete
delete-promptTry in Inspector

Permanently deletes all versions of a prompt. Fails if prompt is used by any agent.

ParametersJSON Schema
NameRequiredDescriptionDefault
prompt_nameYesPrompt name to delete
get-agentTry in Inspector

Get full config for a single agent by ID.

ParametersJSON Schema
NameRequiredDescriptionDefault
agent_idYesAgent UUID
get-agent-first-messageTry in Inspector

Retrieves the agent's greeting message and creates a new session. Call this before starting a chat.

ParametersJSON Schema
NameRequiredDescriptionDefault
agent_idYesAgent UUID
get-chat-sessionTry in Inspector

Retrieves complete chat session including all messages, status, and metadata.

ParametersJSON Schema
NameRequiredDescriptionDefault
session_idYesSession UUID to retrieve
get-conversationTry in Inspector

Retrieves full conversation details including transcript, analysis, and metadata for a specific conversation ID.

ParametersJSON Schema
NameRequiredDescriptionDefault
conversation_idYesConversation UUID, e.g., '550e8400-e29b-41d4-a716-446655440000'
get-mobile-numbersTry in Inspector

Get all purchased mobile numbers with their telephony provider (PLIVO/TWILIO). Use this to find available numbers for outbound calls.

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

get-promptTry in Inspector

Retrieves a prompt by name. Returns latest version unless a specific version is requested.

ParametersJSON Schema
NameRequiredDescriptionDefault
versionNoVersion number, omit for latest
prompt_nameYesPrompt name, e.g., 'customer_support_v1'
get-prompt-versionsTry in Inspector

Retrieves version history for a prompt, sorted by version number descending.

ParametersJSON Schema
NameRequiredDescriptionDefault
prompt_nameYesPrompt name to get versions for
initiate-bulk-outbound-callsTry in Inspector

Initiates multiple outbound voice calls in one request. Each call can have unique metadata.

ParametersJSON Schema
NameRequiredDescriptionDefault
callsYesArray of call objects with phone and optional metadata
agent_idYesAgent UUID to use for all calls
is_test_callNoMark as test calls, default false
telephony_providerNoPLIVO or TWILIO, default PLIVOPLIVO
initiate-outbound-callTry in Inspector

Initiates an outbound voice call using an AI agent. 10-digit numbers are auto-prefixed with +91.

ParametersJSON Schema
NameRequiredDescriptionDefault
agent_idYesAgent UUID to use for the call
metadataNoKey-value data for agent context, e.g., {customer_name: 'John'}
is_test_callNoMark as test call, default false
mobile_numberYesPhone number in E.164 format (+919876543210) or 10-digit Indian number
telephony_providerNoPLIVO or TWILIO, default PLIVOPLIVO
list-agentsTry in Inspector

List voice agents. Returns a paginated list. Default page_size is 10 to keep responses lightweight — only increase if the user explicitly asks for more.

ParametersJSON Schema
NameRequiredDescriptionDefault
pageNoPage number, starts at 1
searchNoFilter by agent name
page_sizeNoResults per page (default 10, max 100). Keep low unless user requests more.
agent_typeNoINBOUND or OUTBOUND
agent_categoryNoAgent category filter
list-chat-sessionsTry in Inspector

List text chat sessions. Default page_size is 10 — only increase if the user explicitly asks for more.

ParametersJSON Schema
NameRequiredDescriptionDefault
pageNoPage number, starts at 1
agent_idNoFilter by agent UUID
page_sizeNoResults per page (default 10, max 100). Keep low unless user requests more.
list-conversationsTry in Inspector

List voice and chat conversations. Default page_size is 10 — only increase if the user explicitly asks for more. Use filters to narrow results.

ParametersJSON Schema
NameRequiredDescriptionDefault
pageNoPage number, starts at 1
searchNoSearch text in conversation metadata
to_dateNoEnd date filter (ISO 8601)
to_timeNoFilter by time of day - end (HH:MM)
agent_idNoFilter by agent ID
from_dateNoStart date filter (ISO 8601)
from_timeNoFilter by time of day - start (HH:MM)
page_sizeNoResults per page (default 10, max 50). Keep low unless user requests more.
calling_typeNoFilter by calling direction (INBOUND/OUTBOUND)
conversation_idNoFilter by specific conversation ID
timezone_offsetNoTimezone offset in minutes from UTC (default: 330 for IST)
conversation_typeNoFilter by conversation type
conversation_statusNoFilter by conversation status
list-promptsTry in Inspector

List prompts. Default page_size is 10 — only increase if the user explicitly asks for more. Returns latest version of each prompt by default.

ParametersJSON Schema
NameRequiredDescriptionDefault
pageNoPage number, starts at 1
searchNoFilter by prompt name
page_sizeNoResults per page (default 10, max 100). Keep low unless user requests more.
latest_onlyNoIf false, returns all versions
list-providersTry in Inspector

Get list of all available providers (LLM, TTS, STT) with their models. Helps understand available options when configuring agents.

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

list-voicesTry in Inspector

List voices for a TTS provider. Default page_size is 20 — only increase if the user explicitly asks for more.

ParametersJSON Schema
NameRequiredDescriptionDefault
pageNoPage number, starts at 1
modelNoFilter by model (e.g., bulbul:v2, bulbul:v3)
genderNoFilter by gender
searchNoSearch by voice name
languageNoFilter by language code (e.g., en, hi, es)
providerYesVoice provider (CARTESIA, ELEVENLABS, SARVAM, GOOGLE, HUME)
page_sizeNoResults per page (default 20, max 100). Keep low unless user requests more.
show_cloned_onlyNoShow only cloned voices (default false)
send-chat-messageTry in Inspector

Sends a message to an AI agent and returns the response. Omit session_id to start a new conversation.

ParametersJSON Schema
NameRequiredDescriptionDefault
messageYesMessage text, 1-4000 characters
agent_idYesAgent UUID to chat with
metadataNoCustom key-value metadata, e.g., {user_id: '123'}
streamingNoUse streaming response, default false
session_idNoExisting session UUID to continue conversation
test-voice-agentTry in Inspector

Opens an interactive UI widget to test an AI agent in real-time. Supports voice, chat, and avatar interactions. Provide an agent ID to launch the testing widget.

ParametersJSON Schema
NameRequiredDescriptionDefault
agent_idYesAgent ID to test interactively, e.g., '68c3c7ceaf78c5e22fcbcd31'
update-agentTry in Inspector

Update an existing agent. Only include fields to change. IMPORTANT: If changing STT, LLM, or TTS config, call 'list-providers' first to validate. If changing voice_id, call 'list-voices' first. If assigning mobile numbers, call 'get-mobile-numbers' first.

ParametersJSON Schema
NameRequiredDescriptionDefault
llmNoNew LLM configuration
sttNoNew STT configuration
ttsNoNew TTS configuration
toolsNoUpdate agent tools. Only specified tools are updated.
promptNoReference existing prompt instead
agent_idYesAgent UUID to update
agent_nameNoNew name, 1-100 characters
agent_tipsNoNew usage tips array
agent_typeNoINBOUND or OUTBOUND
widget_configNoUpdate widget configuration
agent_categoryNoNew category
prompt_contentNoNew prompt text, creates new version
agent_descriptionNoNew description, max 500 characters
agent_mobile_numberNoMobile number for the agent as a single-element array, e.g. ["918035315555"]. Number format: country code + number, no '+' prefix. Call 'get-mobile-numbers' first to get valid purchased numbers. INBOUND agents require unique numbers.
first_message_promptNoNew greeting message
update-promptTry in Inspector

Updates a prompt by creating a new version. Version number auto-increments.

ParametersJSON Schema
NameRequiredDescriptionDefault
prompt_nameYesName of existing prompt to update
prompt_typeNo'main' for agent prompts, 'analysis' for analysis
prompt_contentYesNew prompt text content

FAQ

How do I claim this server?

To claim this server, publish a /.well-known/glama.json file on your server's domain with the following structure:

{ "$schema": "https://glama.ai/mcp/schemas/connector.json", "maintainers": [ { "email": "your-email@example.com" } ] }

The email address must match the email associated with your Glama account. Once verified, the server will appear as claimed by you.

What are the benefits of claiming a server?
  • Control your server's listing on Glama, including description and metadata
  • Receive usage reports showing how your server is being used
  • Get monitoring and health status updates for your server

Discussions

No comments yet. Be the first to start the discussion!

Try in Browser

Your Connectors

Sign in to create a connector for this server.