> ## Documentation Index
> Fetch the complete documentation index at: https://composio-preview.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Lyzr

> Build, run, evaluate, and manage AI agents and related Lyzr Studio services, including voice, responsible AI, simulation, memory, scheduling, workflows, and knowledge bases.

![Lyzr logo](https://mintcdn.com/composio-preview/ufpARLIsio16odJB/images/toolkits/lyzr.svg?fit=max\&auto=format\&n=ufpARLIsio16odJB\&q=85\&s=b1d114c2d44b980b9bd2e1ff2fe73790)

* **Category:** ai agents
* **Auth:** API\_KEY
* **Composio-managed OAuth available?** N/A
* **Tools:** 22
* **Triggers:** 0
* **Slug:** `LYZR`
* **Version:** 20261030\_00

## Tools

> [accordion] Add Memory Messages
>
> **Slug:** `LYZR_ADD_MEMORY_MESSAGES`
>
> Store a batch of conversation messages in Lyzr Memory for an owner, session, and optional agent, with optional synchronous fact extraction.
>
> #### Input Parameters
>
> **agent\_id** `string`
>
> Optional Agent Studio agent ID associated with the conversation.
>
> **messages** `array` *required*
>
> One or more conversation messages to add in this batch.
>
> **owner\_id** `string` *required*
>
> Stable application user or customer ID that owns these memories.
>
> **session\_id** `string` *required*
>
> Conversation session ID used to group the messages.
>
> **instructions** `string`
>
> Optional custom instructions layered onto the memory extraction prompt.
>
> **document\_mode** `boolean`
>
> Whether to store and embed each message verbatim without conversational fact extraction.
>
> **provider\_type** `string`
>
> Memory provider identifier. Use cognis unless an external provider is configured.
>
> **sync\_extraction** `boolean`
>
> Whether to wait for long-term fact extraction before returning.
>
> **memory\_credential\_id** `string`
>
> Existing provider credential ID required only for bring-your-own memory providers.
>
> **extract\_assistant\_facts** `boolean`
>
> Whether assistant messages may also produce extracted facts.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Chat With Agent
>
> **Slug:** `LYZR_CHAT_WITH_AGENT`
>
> Send a message or message history to an active Agent Studio agent and return its synchronous inference response for a session.
>
> #### Input Parameters
>
> **assets** `array`
>
> Optional Agent Studio asset IDs available during this inference call.
>
> **message** `string`
>
> Single user message. Provide this or messages, but not both.
>
> **user\_id** `string`
>
> Stable end-user identifier used by Lyzr for session ownership and memory.
>
> **agent\_id** `string` *required*
>
> Active Agent Studio agent ID to run.
>
> **messages** `array`
>
> Chat history for multi-message inference. Provide this or message, but not both.
>
> **session\_id** `string` *required*
>
> Caller-chosen stable session ID used to continue the conversation.
>
> **filter\_variables** `object`
>
> Optional metadata filters used by the agent's knowledge retrieval.
>
> **system\_prompt\_variables** `object`
>
> Values substituted into variables declared by the agent's system prompt.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Clear Memory Session
>
> **Slug:** `LYZR_CLEAR_MEMORY_SESSION`
>
> Permanently clear messages and memories for one explicit owner and session in Lyzr Memory. Cross-session deletion is always disabled.
>
> #### Input Parameters
>
> **agent\_id** `string`
>
> Optional agent ID used to further constrain the session cleanup.
>
> **owner\_id** `string` *required*
>
> Owner ID whose session data should be permanently cleared.
>
> **session\_id** `string` *required*
>
> Exact session ID to clear. This tool never clears other sessions.
>
> **provider\_type** `string`
>
> Memory provider identifier used by the session.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Create Agent
>
> **Slug:** `LYZR_CREATE_AGENT`
>
> Create a single-task Agent Studio agent with an LLM provider, model, instructions, and optional tool configuration.
>
> #### Input Parameters
>
> **name** `string` *required*
>
> Human-readable name for the new agent.
>
> **tool** `string`
>
> Optional Lyzr tool ID assigned to this single-task agent.
>
> **model** `string` *required*
>
> Provider model identifier the agent should use.
>
> **top\_p** `number`
>
> Nucleus-sampling probability mass.
>
> **is\_active** `boolean`
>
> Whether inference is enabled immediately after creation.
>
> **agent\_goal** `string`
>
> Outcome the agent should optimize for.
>
> **agent\_role** `string`
>
> Role or persona the agent should adopt.
>
> **description** `string`
>
> Short explanation of the agent's purpose.
>
> **provider\_id** `string` *required*
>
> Lyzr LLM provider ID configured in Agent Studio.
>
> **temperature** `number`
>
> Sampling temperature for model responses.
>
> **agent\_output** `string`
>
> Expected output format or content requirements.
>
> **agent\_context** `string`
>
> Background context available on every run.
>
> **max\_iterations** `integer`
>
> Optional maximum reasoning or tool iterations per inference run.
>
> **store\_messages** `boolean`
>
> Whether Agent Studio should retain conversation messages.
>
> **llm\_credential\_id** `string`
>
> Optional existing Agent Studio credential ID for the selected LLM provider.
>
> **agent\_instructions** `string` *required*
>
> Operational instructions defining how the agent should perform its task.
>
> **tool\_usage\_description** `string`
>
> Guidance telling the agent when and how to use its assigned tool.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Create Agent Schedule
>
> **Slug:** `LYZR_CREATE_AGENT_SCHEDULE`
>
> Create a cron schedule that runs one Agent Studio agent with a message in a specified timezone.
>
> #### Input Parameters
>
> **message** `string`
>
> Message supplied to the agent for each scheduled run.
>
> **user\_id** `string` *required*
>
> Stable Lyzr user ID attributed to scheduled executions.
>
> **agent\_id** `string` *required*
>
> Agent Studio agent ID to run on schedule.
>
> **timezone** `string`
>
> IANA timezone used to interpret the cron expression, for example America/New\_York.
>
> **max\_retries** `integer`
>
> Maximum retries after a failed scheduled execution (0-5).
>
> **retry\_delay** `integer`
>
> Delay in seconds between retries (10-3600).
>
> **cron\_expression** `string` *required*
>
> Five-field cron expression in minute, hour, day, month, and weekday order, for example 0 9 \* \* 1.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Delete Agent
>
> **Slug:** `LYZR_DELETE_AGENT`
>
> Permanently delete one Agent Studio agent by ID.
>
> #### Input Parameters
>
> **agent\_id** `string` *required*
>
> Agent Studio agent ID to permanently delete.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Delete Agent Schedule
>
> **Slug:** `LYZR_DELETE_AGENT_SCHEDULE`
>
> Permanently delete one agent cron schedule and remove its associated scheduler job.
>
> #### Input Parameters
>
> **schedule\_id** `string` *required*
>
> Schedule ID returned by Create Agent Schedule or List Agent Schedules.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Evaluate Text With RAI Policy
>
> **Slug:** `LYZR_EVALUATE_TEXT_WITH_RAI_POLICY`
>
> Apply one owned Responsible AI policy to text and return processed text, block decisions, warnings, detections, and redactions.
>
> #### Input Parameters
>
> **query** `string`
>
> Optional originating user query for contextual checks.
>
> **stage** `string ("llm_input" | "llm_output")`
>
> Whether the text is entering or leaving an LLM.
>
> **agent\_id** `string`
>
> Optional Agent Studio agent ID for audit context.
>
> **policy\_id** `string` *required*
>
> RAI policy ID returned by List RAI Policies.
>
> **input\_text** `string` *required*
>
> Text to evaluate and, where configured, redact or transform.
>
> **session\_id** `string`
>
> Optional conversation session ID for audit context.
>
> **grounding\_source** `string`
>
> Optional trusted source text used by groundedness checks.
>
> **agent\_instructions** `string`
>
> Optional agent instructions used when evaluating policy compliance.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Agent
>
> **Slug:** `LYZR_GET_AGENT`
>
> Return the current configuration and status of one Agent Studio agent by ID.
>
> #### Input Parameters
>
> **agent\_id** `string` *required*
>
> Agent Studio agent ID returned by List Agents or Create Agent.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Knowledge Base
>
> **Slug:** `LYZR_GET_KNOWLEDGE_BASE`
>
> Return one Lyzr Knowledge Base configuration by its known RAG configuration ID.
>
> #### Input Parameters
>
> **config\_id** `string` *required*
>
> Known Lyzr RAG configuration ID. Lyzr does not provide a list endpoint for discovering this value.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Simulation Overview
>
> **Slug:** `LYZR_GET_SIMULATION_OVERVIEW`
>
> Return aggregate Agent Simulation counts and average evaluation pass rate for the connected workspace.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Agents
>
> **Slug:** `LYZR_LIST_AGENTS`
>
> Return all Agent Studio agents owned by the connected API key, including IDs needed by lifecycle and inference tools.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Agent Schedules
>
> **Slug:** `LYZR_LIST_AGENT_SCHEDULES`
>
> Return one page of agent cron schedules, optionally filtered by user, agent, or active state.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum schedules to return in this page (1-100).
>
> **user\_id** `string`
>
> Return schedules attributed to this Lyzr user ID.
>
> **agent\_id** `string`
>
> Return schedules for this Agent Studio agent ID.
>
> **is\_active** `boolean`
>
> True for active schedules, false for paused schedules, or omit for both.
>
> **next\_cursor** `string`
>
> Continuation cursor returned by a previous call. Omit for the first page.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Agent Simulation Issues
>
> **Slug:** `LYZR_LIST_AGENT_SIMULATION_ISSUES`
>
> Return recent performance, error, and Responsible AI issues across monitored agents, optionally filtered by resolution and severity.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum recent issues to return (1-100).
>
> **resolved** `boolean`
>
> True for acknowledged issues, false for unresolved issues, or omit for both.
>
> **severity** `string ("critical" | "medium" | "low")`
>
> Restrict results to one severity. Critical also includes legacy high-severity issues.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List RAI Policies
>
> **Slug:** `LYZR_LIST_RAI_POLICIES`
>
> Return Responsible AI policies owned by the authenticated user, including policy IDs and enabled guardrail configuration.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List SuperFlow Executions
>
> **Slug:** `LYZR_LIST_SUPERFLOW_EXECUTIONS`
>
> Return one page of SuperFlow execution records, optionally filtered by workflow and execution status.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum executions to return in this page (1-200).
>
> **status** `string ("running" | "completed" | "failed" | "cancelled" | "paused")`
>
> Restrict results to one execution status.
>
> **next\_cursor** `string`
>
> Offset cursor returned by a previous call. Omit for the first page.
>
> **workflow\_id** `string`
>
> Return executions for this SuperFlow workflow ID.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List SuperFlow Workflows
>
> **Slug:** `LYZR_LIST_SUPERFLOW_WORKFLOWS`
>
> Return all SuperFlow workflow definitions accessible to the connected API key, including workflow IDs and graph data.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Voice Agents
>
> **Slug:** `LYZR_LIST_VOICE_AGENTS`
>
> Return saved Voice agents accessible to the connected organization, including IDs used to filter transcripts.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Voice Transcripts
>
> **Slug:** `LYZR_LIST_VOICE_TRANSCRIPTS`
>
> Return one page of accessible Voice call transcripts, optionally filtered by agent, session, organization, or date range.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum transcripts to return in this page (1-200).
>
> **to\_time** `string`
>
> Inclusive ISO date or ISO date-time upper bound.
>
> **agent\_id** `string`
>
> Return transcripts for this Voice agent ID.
>
> **from\_time** `string`
>
> Inclusive ISO date or ISO date-time lower bound.
>
> **session\_id** `string`
>
> Return only the transcript for this Voice session ID.
>
> **sort\_order** `string ("asc" | "desc")`
>
> Sort transcripts by time ascending or descending.
>
> **next\_cursor** `string`
>
> Continuation cursor returned by a previous call. Omit for the first page.
>
> **organization\_id** `string`
>
> Restrict results to an accessible organization ID.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Retrieve Knowledge
>
> **Slug:** `LYZR_RETRIEVE_KNOWLEDGE`
>
> Retrieve relevant documents from a known Lyzr Knowledge Base using configurable basic, MMR, hypothetical-document, or time-aware retrieval.
>
> #### Input Parameters
>
> **query** `string` *required*
>
> Natural-language query used to rank stored documents.
>
> **top\_k** `integer`
>
> Maximum relevant documents to return (1-100).
>
> **retrieval\_type** `string ("basic" | "mmr" | "hyde" | "time_aware")`
>
> Retrieval strategy: basic, MMR, hypothetical-document, or time-aware.
>
> **score\_threshold** `number`
>
> Minimum document relevance score from 0.0 to 1.0.
>
> **knowledge\_base\_id** `string` *required*
>
> Known 24-character hexadecimal Lyzr RAG configuration ID to query.
>
> **time\_decay\_factor** `number`
>
> Non-negative weight applied to document recency for time-aware retrieval.
>
> **mmr\_relevance\_diversity\_balance** `number`
>
> MMR balance from 0.0 to 1.0 between query relevance and result diversity. Used only when retrieval\_type is mmr; keep the default 0.5 unless retrieval behavior has been calibrated.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Search Memories
>
> **Slug:** `LYZR_SEARCH_MEMORIES`
>
> Semantically search stored memories for one owner, within one session or across that owner's sessions, using Lyzr's hybrid vector and BM25 retrieval.
>
> #### Input Parameters
>
> **limit** `integer`
>
> Maximum matching memories to return (1-100).
>
> **query** `string` *required*
>
> Natural-language memory search query.
>
> **agent\_id** `string`
>
> Optional Agent Studio agent ID filter.
>
> **category** `string`
>
> Optional provider-defined memory category filter.
>
> **owner\_id** `string` *required*
>
> Stable application user or customer ID whose memories should be searched.
>
> **session\_id** `string`
>
> Session whose memories should be searched. Required unless cross\_session is true.
>
> **cross\_session** `boolean`
>
> Whether to search across all sessions belonging to the owner.
>
> **provider\_type** `string`
>
> Memory provider identifier used when the memories were stored.
>
> **include\_historical** `boolean`
>
> Whether archived or superseded memories may be returned.
>
> **memory\_credential\_id** `string`
>
> Existing provider credential ID required only for bring-your-own memory providers.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Set Agent Status
>
> **Slug:** `LYZR_SET_AGENT_STATUS`
>
> Activate or deactivate an Agent Studio agent. Deactivated agents reject inference until reactivated.
>
> #### Input Parameters
>
> **agent\_id** `string` *required*
>
> Agent Studio agent ID to activate or deactivate.
>
> **is\_active** `boolean` *required*
>
> True to allow inference, or false to reject inference while preserving the agent.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not
