> ## 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.

# Lakera Guard

> Lakera Guard screens AI application interactions for security and safety risks and provides Enterprise APIs for managing guardrail policies and projects.

![Lakera Guard logo](https://mintcdn.com/composio-preview/tIuAovF9-7pXq2wA/images/toolkits/lakera_guard.svg?fit=max\&auto=format\&n=tIuAovF9-7pXq2wA\&q=85\&s=1943f3c32fc02c5aa4c1dc988b07fafb)

* **Category:** ai safety compliance detection
* **Auth:** API\_KEY
* **Composio-managed OAuth available?** N/A
* **Tools:** 10
* **Triggers:** 0
* **Slug:** `LAKERA_GUARD`
* **Version:** 20261101\_00

## Tools

> [accordion] Create Policy
>
> **Slug:** `LAKERA_GUARD_CREATE_POLICY`
>
> Create an Enterprise Lakera policy with built-in detectors and/or custom guardrails. Lakera may omit the new policy ID; if omitted, later retrieval, replacement, and deletion require external policy lookup.
>
> #### Input Parameters
>
> **name** `string` *required*
>
> Policy name, unique within the organization.
>
> **detectors** `array`
>
> Built-in detector configurations. Provide at least one detector or custom\_guardrail\_id.
>
> **default\_threshold** `string ("l1_confident" | "l2_very_likely" | "l3_likely" | "l4_less_likely")` *required*
>
> Default confidence threshold at or above which configured detectors trigger.
>
> **custom\_guardrail\_ids** `array`
>
> Existing custom guardrail IDs to include. Provide at least one ID or detector.
>
> #### 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 Project
>
> **Slug:** `LAKERA_GUARD_CREATE_PROJECT`
>
> Create an Enterprise Lakera project, optionally assigning a policy, metadata, and detect-or-flag action. Omit policy\_id to use the default policy. Lakera does not return the generated project ID, so later retrieval, updates, and deletion require external project lookup.
>
> #### Input Parameters
>
> **name** `string` *required*
>
> Name for the new project.
>
> **action** `string ("detect" | "flag")`
>
> Project response action: detect reports detections, while flag applies flagging behavior.
>
> **policy\_id** `string`
>
> Existing policy ID to assign; omit to use the default policy.
>
> **project\_metadata** `object`
>
> String-valued project metadata stored by Lakera.
>
> #### 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 Policy
>
> **Slug:** `LAKERA_GUARD_DELETE_POLICY`
>
> Permanently delete an Enterprise Lakera policy by ID. The provider rejects deletion while any project is assigned to the policy.
>
> #### Input Parameters
>
> **policy\_id** `string` *required*
>
> ID of the unassigned policy 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 Project
>
> **Slug:** `LAKERA_GUARD_DELETE_PROJECT`
>
> Permanently and irreversibly delete an Enterprise Lakera project by ID. Historical requests lose their association with the project's metadata and assigned policy.
>
> #### Input Parameters
>
> **project\_id** `string` *required*
>
> ID of the project to permanently and irreversibly 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] Evaluate Detectors
>
> **Slug:** `LAKERA_GUARD_EVALUATE_DETECTORS`
>
> Evaluate content and return one confidence result per configured detector without producing an aggregate flagging decision or Guard screening log. Use for analysis and tuning, not runtime allow/block decisions.
>
> #### Input Parameters
>
> **dev\_info** `boolean`
>
> Whether to request Guard build and model version information.
>
> **messages** `array` *required*
>
> OpenAI-style conversation in chronological order; each message requires a role, while assistant tool-call messages may omit content.
>
> **metadata** `object`
>
> Optional arbitrary request metadata passed to Lakera Guard.
>
> **project\_id** `string`
>
> Lakera project whose policy should be evaluated; omit to use the default policy.
>
> #### 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 Policy
>
> **Slug:** `LAKERA_GUARD_GET_POLICY`
>
> Retrieve an Enterprise Lakera policy by ID, including its detector configuration, custom guardrails, version, and assigned projects.
>
> #### Input Parameters
>
> **policy\_id** `string` *required*
>
> Policy ID returned by Create Policy.
>
> #### 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 Project
>
> **Slug:** `LAKERA_GUARD_GET_PROJECT`
>
> Retrieve an Enterprise Lakera project by ID, including its assigned policy, metadata, and detect-or-flag action.
>
> #### Input Parameters
>
> **project\_id** `string` *required*
>
> Lakera project ID to retrieve.
>
> #### 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] Screen Content
>
> **Slug:** `LAKERA_GUARD_SCREEN_CONTENT`
>
> Screen the latest LLM or agent interaction with Lakera Guard to inform runtime allow/block decisions. Earlier messages provide context but are not re-screened. The result follows the project policy and mode: Detect mode forces the top-level flagged result to false even when detectors find threats.
>
> #### Input Parameters
>
> **payload** `boolean`
>
> Whether to include matched text locations for supported detectors such as PII, profanity, and custom regex.
>
> **dev\_info** `boolean`
>
> Whether to include Guard build and model version information.
>
> **messages** `array` *required*
>
> OpenAI-style conversation in chronological order. Earlier messages provide context, but Guard screens only the latest interaction.
>
> **metadata** `object`
>
> Optional request tags, commonly user\_id, session\_id, ip\_address, and internal\_request\_id.
>
> **breakdown** `boolean`
>
> Whether to include detector-by-detector decisions and confidence levels.
>
> **project\_id** `string`
>
> Lakera project whose policy should screen the interaction; omit to use the default policy.
>
> #### 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
>
> **Slug:** `LAKERA_GUARD_UPDATE_POLICY`
>
> Replace an Enterprise Lakera policy's complete configuration and create a new policy version. Omitted detectors or custom guardrails are removed, so retrieve the policy first when preserving existing settings.
>
> #### Input Parameters
>
> **name** `string` *required*
>
> Complete replacement policy name.
>
> **detectors** `array`
>
> Complete replacement built-in detector list. Omission removes existing detectors; provide at least one detector or custom guardrail ID.
>
> **policy\_id** `string` *required*
>
> ID of the policy to replace.
>
> **default\_threshold** `string ("l1_confident" | "l2_very_likely" | "l3_likely" | "l4_less_likely")` *required*
>
> Complete replacement default detector threshold.
>
> **custom\_guardrail\_ids** `array`
>
> Complete replacement custom guardrail ID list. Omission removes existing custom guardrails.
>
> #### 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 Project
>
> **Slug:** `LAKERA_GUARD_UPDATE_PROJECT`
>
> Update an Enterprise Lakera project by ID. Name is required by the official request schema; retrieve the project first when preserving its current name or other optional settings.
>
> #### Input Parameters
>
> **name** `string` *required*
>
> Project name; required by the official Platform request schema.
>
> **action** `string ("detect" | "flag")`
>
> Project action. Omission keeps the existing action.
>
> **policy\_id** `string`
>
> Policy ID to assign to the project.
>
> **project\_id** `string` *required*
>
> ID of the project to update.
>
> **project\_metadata** `object`
>
> String-valued project metadata to send in the update.
>
> #### 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
