Relevance AI · AsyncAPI Specification
Relevance Ai Events
Version
View Spec
View on GitHub
AI AgentsAI AutomationMulti-Agent SystemsAgent PlatformMCPagent-nativeWorkflow-AutomationLLM OrchestrationKnowledge-ManagementObservabilitySales AutomationGTM EngineeringAsyncAPIEvents
AsyncAPI Specification
generated: '2026-08-29'
method: searched
source: >-
https://relevanceai.com/docs/enterprise/streaming-events,
https://relevanceai.com/docs/build/agents/build-your-agent/agent-triggers/custom-webhook,
https://relevanceai.com/docs/build/agents/build-your-agent/triggers,
openapi/relevance-ai-openapi.json, https://github.com/RelevanceAI/relevance-js-sdk
provider: Relevance AI
providerId: relevance-ai
description: >-
Relevance AI has three distinct event surfaces and publishes no AsyncAPI for any of them.
(1) INBOUND webhook triggers, where an external system POSTs to a per-agent URL to start the
agent. (2) An OUTBOUND OpenTelemetry event feed — audit logs and execution traces delivered
to the customer's own S3 bucket, Enterprise only. (3) A server-sent-events stream for live
task output. Nothing here was fabricated; where a shape is not published, it says so.
asyncapi_spec:
published: false
probed:
- url: https://relevanceai.com/asyncapi.yaml
note: not present (site 404 page)
- url: https://api-f1db6c.stack.tryrelevance.com/latest/asyncapi.json
note: not present
note: >-
N/A rather than a defect for the OTEL feed, which is an OpenTelemetry contract rather than
an AsyncAPI one. The webhook and SSE surfaces genuinely have no machine-readable contract.
surfaces:
- id: webhook-triggers
direction: inbound
type: webhook
description: >-
Each agent can expose a unique webhook URL. Any system that can send an HTTP request can
POST to it; the payload becomes a message to the agent.
endpoint: per-agent unique URL, issued in the agent's Triggers panel
payload: >-
Arbitrary caller-defined JSON. A message template maps it into the agent prompt with
double-brace variables ({{first_name}}), or {{$}} passes the whole payload.
configuration:
- name: message_template
detail: 'Template variables in double braces; {{$}} for the entire payload.'
- name: thread_id_mapping
detail: Maps a payload field to a Thread ID so related messages share a conversation.
- name: unique_id_mapping
detail: >-
Maps a payload field used to suppress duplicate processing. This is the only
deduplication mechanism on the surface — there is no signature header.
security:
signature: none documented
detail: >-
The provider's own FAQ says security rests on the URL being "difficult to guess", plus
optional unique-ID deduplication. No HMAC signature, no shared secret, no mTLS is
documented. Treat the URL as a bearer credential.
controls:
- Pause / resume individual triggers
- Pause all / resume all
related_operations:
- TriggerStudioFromWebhook
- SlackAgentsWebhook
source: https://relevanceai.com/docs/build/agents/build-your-agent/agent-triggers/custom-webhook
- id: otel-event-streaming
direction: outbound
type: telemetry-feed
standard: OpenTelemetry
availability: Enterprise plan only
transport: gzipped OTEL JSON objects written to the customer's own S3 bucket
collector_required: false
destinations_supported: [s3]
roadmap: 'Direct OTEL collector endpoints are stated as "on our roadmap".'
path_pattern: >-
{prefix}/customer-otel-{type}-formatted/org_id={org_id}/dt={YYYY-MM-DD}/year={YYYY}/month={MM}/day={DD}/hour={HH}/minute={MM}/{type}_{org_id}_{timestamp}_{uuid}.json.gz
regions:
Australia: ap-southeast-2
Europe: eu-west-2
US: us-east-1
staging_retention_days: 30
source: https://relevanceai.com/docs/enterprise/streaming-events
event_types:
- name: logs
description: Audit logs for administrative and lifecycle events.
- name: traces
description: Execution traces for agents, workforces and LLM completions.
base_attributes:
- relevance_ai.organization_id
- relevance_ai.project_id
- relevance_ai.user_id
- relevance_ai.user_email
- relevance_ai.user_type
- relevance_ai.ip_address
- relevance_ai.device_info
events:
- name: agent_created
category: agent
attributes: [relevance_ai.event.agent_id, relevance_ai.event.cloned_from_agent_id, relevance_ai.event.cloned_from_region, relevance_ai.event.cloned_from_project_id]
- name: agent_updated
category: agent
attributes: [relevance_ai.event.agent_id]
- name: agent_deleted
category: agent
attributes: [relevance_ai.event.agent_id]
- name: agent_draft_saved
category: agent
attributes: [relevance_ai.event.agent_id, relevance_ai.event.version_id]
- name: agent_published
category: agent
attributes: [relevance_ai.event.agent_id, relevance_ai.event.version_id]
- name: tool_created
category: tool
attributes: [relevance_ai.event.tool_id, relevance_ai.event.cloned_from_tool_id, relevance_ai.event.cloned_from_region, relevance_ai.event.cloned_from_project_id]
- name: tool_deleted
category: tool
attributes: [relevance_ai.event.tool_id]
- name: tool_draft_saved
category: tool
attributes: [relevance_ai.event.tool_id, relevance_ai.event.version_id]
- name: tool_published
category: tool
attributes: [relevance_ai.event.tool_id, relevance_ai.event.version_id]
- name: workforce_created
category: workforce
attributes: [relevance_ai.event.workforce_id, relevance_ai.event.cloned_from_workforce_id, relevance_ai.event.cloned_from_region, relevance_ai.event.cloned_from_project_id]
- name: workforce_deleted
category: workforce
attributes: [relevance_ai.event.workforce_id]
- name: workforce_draft_saved
category: workforce
attributes: [relevance_ai.event.workforce_id, relevance_ai.event.version_id]
- name: workforce_published
category: workforce
attributes: [relevance_ai.event.workforce_id, relevance_ai.event.version_id]
- name: project_user_role_updated
category: permission
attributes: [relevance_ai.event.target_user_id, relevance_ai.event.project_role]
- name: organization_user_role_updated
category: permission
attributes: [relevance_ai.event.target_user_id, relevance_ai.event.organization_role]
event_count: 15
spans:
- name: invoke_agent
semantic_convention: OTEL GenAI
key_attributes:
- gen_ai.operation.name
- gen_ai.agent.id
- gen_ai.agent.name
- gen_ai.conversation.id
- relevance_ai.agent.relevance_model
- relevance_ai.agent.runtime
- relevance_ai.agent.version_id
- relevance_ai.prompt_injection.detected
- relevance_ai.prompt_injection.confidence
- name: chat
semantic_convention: OTEL GenAI
key_attributes:
- gen_ai.request.model
- gen_ai.request.temperature
- gen_ai.request.max_tokens
- gen_ai.input.messages
- gen_ai.output.messages
- gen_ai.tool.definitions
- gen_ai.response.finish_reasons
- gen_ai.usage.input_tokens
- name: multi_agent_system_trigger
semantic_convention: OTEL GenAI
- name: condition_trigger
semantic_convention: OTEL GenAI
span_count: 4
resource_attributes:
service.name: Relevance AI
privacy:
pii_redaction:
availability: Enterprise, enabled by the Relevance AI team (no self-serve UI)
engine: Presidio-powered ML detection
always_scrubbed: [user email addresses, IP addresses, device info]
conditional_scrubbed: [LLM input messages, LLM output messages, system instructions]
entities: [EMAIL_ADDRESS, PHONE_NUMBER, PERSON, CREDIT_CARD, US_SSN, IP_ADDRESS, IBAN_CODE, URL, DATE_TIME, LOCATION]
actions: [replace, mask, redact, hash]
default_action: replace
- id: sse-streaming
direction: outbound
type: server-sent-events
endpoint: https://{region}.streaming.tryrelevance.com/v1/stream
auth: 'bearer token passed as the ?authorization query parameter'
description: >-
Live task/message streaming, constructed by the SDK's regionStreamingURL(). A related
REST operation, GetWorkforceStreamingToken, mints the workforce streaming token.
contract: >-
No published event schema. The endpoint is documented only in SDK source and the
streaming SDK guide; the message shapes are not specified anywhere machine-readable.
source: https://github.com/RelevanceAI/relevance-js-sdk/blob/main/region.ts
gaps:
- No AsyncAPI document for any of the three surfaces.
- Inbound webhooks carry no signature or shared secret — only an unguessable URL.
- The SSE stream has no published event schema.
- The OTEL feed is Enterprise-only and S3-only; no push endpoint for other tiers.
Work with this as data
Every AsyncAPI spec here is available over the APIs.io API and to AI agents over MCP.
MCP server
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
Tools for asyncapi
4 MCP tools reach this
find_asyncapisBrowse and filter every AsyncAPI spec in the catalog.apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.resolveTurn a domain, URL or GitHub org into the provider it belongs to.find_cohortsEvery scored population of providers in the catalog.
Call it yourself
curl for this page
This AsyncAPI spec
curl "https://apis.io/api/v1/asyncapis/relevance-ai-events"
All asyncapi
curl "https://apis.io/api/v1/asyncapis?limit=25"
Discovery needs no key. Ratings and market analysis are Pro.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.
A second provider on the same verified email joins the account you already have.