> ## 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.

# Metatextai

> Metatext AI specializes in natural language processing and text generation, helping organizations automate writing tasks, sentiment analysis, or content moderation

![Metatextai logo](https://mintcdn.com/composio-preview/9YPuzjEU4UlMnwGq/images/toolkits/metatextai.svg?fit=max\&auto=format\&n=9YPuzjEU4UlMnwGq\&q=85\&s=68b2c861fbfda57d41efb0ebe4fd639b)

* **Category:** artificial intelligence
* **Auth:** API\_KEY
* **Composio-managed OAuth available?** N/A
* **Tools:** 13
* **Triggers:** 0
* **Slug:** `METATEXTAI`
* **Version:** 20261101\_00

## Tools

> [accordion] Chat Completions
>
> **Slug:** `METATEXTAI_CHAT_COMPLETIONS`
>
> Generate OpenAI-compatible chat completions for a chosen model from a prompt or conversation messages. Control generation with temperature, a stop sequence, and a maximum token count.
>
> #### Input Parameters
>
> **stop** `string`
>
> Optional stop sequence to end generation when encountered.
>
> **model** `string` *required*
>
> Model to use for chat completion.
>
> **prompt** `string`
>
> Optional prompt text; alternative to using 'messages'.
>
> **project** `string`
>
> Optional project identifier for billing or project-specific settings.
>
> **messages** `array`
>
> List of message objects for conversation; alternative to 'prompt'.
>
> **max\_tokens** `integer`
>
> Maximum number of tokens to generate.
>
> **temperature** `number`
>
> Sampling temperature to use, between 0.0 and 2.0.
>
> #### 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] Classify Text
>
> **Slug:** `METATEXTAI_CLASSIFY`
>
> Classify text using a project model and return labels with confidence scores. Missing confidence is reported as 0.0; an empty prediction list can also result from access, availability, rate-limit, or network failures.
>
> #### Input Parameters
>
> **text** `string` *required*
>
> Text to classify.
>
> **model** `string`
>
> Model identifier to use. Defaults to the project's primary model if omitted.
>
> **options** `object`
>
> Optional settings to refine classification behavior.
>
> **project\_id** `string` *required*
>
> Identifier of the project to use for classification.
>
> #### 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 Policy Guardrails
>
> **Slug:** `METATEXTAI_CREATE_POLICY_GUARDRAILS`
>
> Create a policy guardrail for a specific application to define automated content rules. Specify the policy's rules and targets, with an optional description and custom violation response.
>
> #### Input Parameters
>
> **id** `string` *required*
>
> Unique identifier for the new policy
>
> **rules** `array` *required*
>
> List of policy rules to enforce
>
> **target** `string` *required*
>
> Where the policy applies. Allowed values: 'user', 'assistant', 'context', 'system'. Aliases: 'output'=assistant, 'input'=\['user','context']
>
> **definition** `string`
>
> Optional short description of the policy
>
> **application\_id** `string` *required*
>
> Identifier of the application in which to create the policy
>
> **override\_response** `string`
>
> Optional custom message returned when a policy violation occurs
>
> #### 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 Guardrail Policy
>
> **Slug:** `METATEXTAI_DELETE_POLICY_GUARDRAILS`
>
> Delete a guardrail policy by application and policy ID. A success result can also reflect a 404 or 410 response and does not necessarily confirm that a policy was deleted.
>
> #### Input Parameters
>
> **policy\_id** `string` *required*
>
> The ID of the policy to delete.
>
> **application\_id** `string` *required*
>
> The ID of the application.
>
> #### 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 Messages
>
> **Slug:** `METATEXTAI_EVALUATE`
>
> Tool to evaluate LLM messages against policies/guardrails. Use after generating model output to get violation details or corrections.
>
> #### Input Parameters
>
> **messages** `array` *required*
>
> Conversation messages to evaluate. Must include 'user' and 'assistant' messages; 'system' optional.
>
> **policies** `array`
>
> Inline policies to evaluate; overrides console defaults when provided.
>
> **fail\_fast** `boolean`
>
> If true, stops evaluation on first violation.
>
> **policy\_ids** `array`
>
> List of policy IDs to override default console policies.
>
> **application** `string`
>
> Application identifier; defaults to your configured application ID.
>
> **override\_response** `string`
>
> Top-level override response message on violation. Overrides correction.
>
> **correction\_enabled** `boolean`
>
> If true, returns corrected output when violations occur (unless override\_response is set).
>
> #### 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] Run Extraction
>
> **Slug:** `METATEXTAI_EXTRACT`
>
> Extract structured data from text using a project model. Empty predictions can also result from access, endpoint availability, or rate-limit failures.
>
> #### Input Parameters
>
> **text** `string` *required*
>
> Text to perform information extraction on
>
> **model** `string`
>
> Optional model identifier to use for extraction
>
> **options** `object`
>
> Additional options for extraction
>
> **project\_id** `string` *required*
>
> Identifier of the project to run extraction on
>
> #### 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] Generate Text
>
> **Slug:** `METATEXTAI_GENERATE`
>
> Generate text for a project model using either a prompt or chat messages. Set temperature, a stop sequence, and a maximum token count to control generation.
>
> #### Input Parameters
>
> **stop** `string`
>
> Sequence at which to stop generating further tokens.
>
> **model** `string` *required*
>
> Model to use for generation (e.g., 'gpt-3.5').
>
> **prompt** `string`
>
> Prompt text for completion. Provide this OR `messages`, not both.
>
> **messages** `array`
>
> List of chat messages for chat-based models. Provide this OR `prompt`, not both.
>
> **max\_tokens** `integer`
>
> Maximum number of tokens to generate.
>
> **project\_id** `string` *required*
>
> Identifier of the project to use for generation.
>
> **temperature** `number`
>
> Sampling temperature between 0 and 2.
>
> #### 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 Applications
>
> **Slug:** `METATEXTAI_LIST_APPLICATIONS`
>
> List MetaTextAI applications to discover their available IDs, names, and descriptions.
>
> #### Input Parameters
>
> **tag** `string`
>
> Filter by tag or label (if supported)
>
> **limit** `integer`
>
> Maximum number of applications to return (if supported)
>
> **offset** `integer`
>
> Number of items to skip (if supported)
>
> **search** `string`
>
> Filter applications by name/description (if supported)
>
> #### 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 Models
>
> **Slug:** `METATEXTAI_LIST_MODELS`
>
> List available models and their supported tasks to help choose a model for chat completions. An empty list can also indicate an authorization or endpoint failure, or an unrecognized response shape.
>
> #### 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 Guardrail Policies
>
> **Slug:** `METATEXTAI_LIST_POLICIES_GUARDRAILS`
>
> List guardrail policies for a specific application using its application ID, to inspect its configured policies.
>
> #### Input Parameters
>
> **application\_id** `string` *required*
>
> Identifier of the application to list policies for
>
> #### 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 Red Team Test Probes
>
> **Slug:** `METATEXTAI_LIST_RED_TEAM_TEST_PROBES`
>
> List available red-team test probe identifiers to discover probes for red teaming. No parameters are required.
>
> #### 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] Run Red Team Test Scan
>
> **Slug:** `METATEXTAI_RUN_RED_TEAM_TEST_SCAN`
>
> Tool to run a vulnerability red-team test scan. Use when you need to execute probes against an application.
>
> #### Input Parameters
>
> **probes** `array`
>
> List of probe identifiers to run. If not provided, default probes will be used.
>
> **application** `string` *required*
>
> The application ID to scan.
>
> #### 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] Update Policy Guardrails
>
> **Slug:** `METATEXTAI_UPDATE_POLICY_GUARDRAILS`
>
> Update an existing application's policy guardrails and rules. The returned policy may echo the request even when all endpoints fail and one returned 403 or 404, so it does not by itself confirm the update.
>
> #### Input Parameters
>
> **id** `string` *required*
>
> Unique policy ID. Must be identical to the path `policy_id`.
>
> **rules** `array` *required*
>
> List of rule objects that define the guardrails.
>
> **target** `string` *required*
>
> Where rules apply. Allowed values: 'user', 'assistant', 'context', 'system'. Alias 'input' maps to \['user','context'], 'output' maps to 'assistant'.
>
> **policy\_id** `string` *required*
>
> The policy ID to update. Must match `id` in the body.
>
> **definition** `string`
>
> Short description of the policy (optional).
>
> **application\_id** `string` *required*
>
> The application ID that scopes the policy.
>
> **override\_response** `string`
>
> Custom message returned when violations occur (optional).
>
> #### 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
