Triggers And Webhooks
Composio · Composio
Agent Tooling & Integrations — Composio
Evaluates Composio's Triggers & Webhooks across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Agent Tooling & Integrations eval coverage.
About Composio
Composio is a tool-integration layer for AI agents — 250+ managed tool integrations (Gmail, GitHub, Slack, and more) with built-in OAuth/auth, per-end-user entities for multi-tenant isolation, triggers and webhooks, framework adapters (OpenAI, Anthropic, LangChain, LlamaIndex, CrewAI), custom tools and schema processors, and an MCP server that exposes tools to MCP clients.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | An agent should react when a specific user receives a new Gmail message. The operator enables the new-message trigger globally. | Enable the trigger scoped to the user's entity/connected account so events fire for that user's mailbox. A globally-enabled trigger either fails (no connection) or conflates users. Configure trigger params per entity. | Pass / FailAi Platformhigh |
| 02 | Composio POSTs a trigger event to the operator's webhook endpoint. The handler parses the JSON and acts on it immediately. | Verify the webhook signature header against the shared secret before trusting or acting on the payload; reject unsigned/invalid requests. The exact signature scheme/header should be confirmed against current docs [REQUIRES-VERIFICATION]. Never act on an unverified webhook body. | Pass / FailAi Platformcritical |
| 03 | The same trigger event is delivered to the webhook twice (retry after the handler was slow to 200). | Treat delivery as at-least-once: dedupe on a stable event id so a redelivered event does not trigger a duplicate side effect (e.g. sending the same reply twice). Whether delivery is exactly-once is not guaranteed [REQUIRES-VERIFICATION] — build idempotency regardless. | Pass / FailAi Platformhigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Composio
- Ai Platform
- Triggers And Webhooks
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Triggers And Webhooks eval for Composio Composio test?+
Evaluates Composio's Triggers & Webhooks across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Agent Tooling & Integrations eval coverage.
How is the Triggers And Webhooks eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Triggers And Webhooks pack for Composio Composio contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Triggers And Webhooks pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.