> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tracectrl.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Guardrail Invocations

> Recent guardrail evaluation spans

Recent `tracectrl.guardrail.evaluation` spans for a specific `(agent_id, guardrail_name)` pair, ordered newest first. Reads directly from `otel_traces` and does not depend on the analytics pipeline tick.

### Query Parameters

**agent\_id** `string` *required*

The agent that owns the guardrail.

**guardrail\_name** `string` *required*

The guardrail to query.

**limit** `integer` (default: 50)

Maximum rows to return.

Returns `400` if `agent_id` or `guardrail_name` is missing.

### Response

Array of `GuardrailInvocation` objects:

**trace\_id** `string`

Trace ID of the evaluation span

**span\_id** `string`

Span ID of the evaluation span

**observed\_at** `string`

ISO-8601 timestamp

**decision** `string`

`pass`, `fail`, or `error`

**timing** `string`

`pre_input` or `post_output`

**reason** `string`

Judge's reasoning

**evidence** `string`

Evidence captured at evaluation time

**severity** `string`

Severity level

**provider** `string`

`judge_llm` or `protector_plus`

**judge\_model** `string`

Model used for the decision

**response\_json** `string`

Structured response payload (Protector Plus only; empty for legacy judge\_llm spans)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.