← Leaderboard
7.1 L3

Knock

Ready Assessed · Docs reviewed ยท Mar 16, 2026 Confidence 0.53 Last evaluated Mar 16, 2026

Score breakdown

Dimension Score Bar
Execution Score

Measures reliability, idempotency, error ergonomics, latency distribution, and schema stability.

7.5
Access Readiness Score

Measures how easily an agent can onboard, authenticate, and start using this service autonomously.

6.4
Aggregate AN Score

Composite score: 70% execution + 30% access readiness.

7.1

Autonomy breakdown

P1 Payment Autonomy
โ€”
G1 Governance Readiness
โ€”
W1 Web Agent Accessibility
โ€”
Overall Autonomy
Pending

Active failure modes

No active failure modes reported.

Reviews

Published review summaries with trust provenance attached to each card.

How are reviews sourced?

Docs-backed Built from public docs and product materials.

Test-backed Backed by guided testing or evaluator-run checks.

Runtime-verified Verified from authenticated runtime evidence.

Knock: Auth & Environment Keys

Docs-backed

API keys are per-environment (development, production). Secret keys for server-side API calls (workflow triggers, user management). Publishable keys for client-side SDK access (feed reads, preference updates). Secret keys are passed via Authorization: Bearer header. The secret/publishable key split ensures client-side code can't trigger workflows. Environment isolation prevents development triggers from reaching production channels. User tokens (signed JWTs) authenticate individual users for client-side operations. For agents, using secret keys for server-side workflow triggers and publishable keys for client-side feed access provides appropriate separation. No OAuth for third-party integrations. Token signing uses the secret key.

Rhumb editorial team Mar 16, 2026

Knock: Comprehensive Agent-Usability Assessment

Docs-backed

Knock provides notification infrastructure that separates notification logic from application code. For agents, the model is: define notification workflows (multi-channel sequences with conditions, delays, and batching), then trigger workflows via API when events occur. Knock handles channel routing (email, push, in-app, SMS, Slack, Discord, WhatsApp), user preference management (users choose which channels they receive notifications on), and delivery orchestration. The key agent value: notification logic (which channels, what delays, batch/digest rules) is managed in Knock's dashboard, not in application code. This means agents can trigger a workflow and Knock handles the complexity of multi-channel, preference-respecting delivery. In-app notification feed components provide pre-built UI. For agents building products with sophisticated notification requirements, Knock reduces notification engineering to a single API call per event.

Rhumb editorial team Mar 16, 2026

Knock: API Design โ€” Workflows, Users & Feeds

Docs-backed

REST API at api.knock.app/v1/. Workflow triggering: POST /workflows/{key}/trigger with recipients (user IDs), actor (who caused the event), and data (template variables). User management: PUT /users/{id} with properties and channel data (push tokens, email addresses). Preferences: GET/PUT /users/{id}/preferences for channel and workflow preferences. In-app feed: GET /users/{id}/feeds/{feed_id}/items returns notification items for the in-app feed. Bulk operations: trigger workflows for multiple recipients in a single request. Object-based triggers enable notifications about entities (comments on a document, updates to a project). Tenant support enables multi-tenant notification scoping. The API is clean and well-designed โ€” workflow triggering is a single endpoint, and Knock handles the orchestration complexity.

Rhumb editorial team Mar 16, 2026

Knock: Error Handling & Delivery Observability

Docs-backed

API errors return JSON with code and message. Standard HTTP status codes. Workflow execution errors are visible in the Knock dashboard with per-step status (sent, skipped, failed) and channel provider error messages. Rate limits are generous for workflow triggers. Message delivery status is tracked per-channel: sent, delivered, read, seen, archived. Batch and digest workflows aggregate notifications over configurable time windows โ€” reducing notification noise. The debug log in the dashboard shows every workflow execution with step-by-step detail. Channel provider failures (SendGrid down, APNs error) are reported with the provider's error message. For agents, the workflow execution visibility eliminates black-box notification debugging โ€” every step's outcome is inspectable.

Rhumb editorial team Mar 16, 2026

Knock: Documentation & Integration Guides

Docs-backed

Documentation at docs.knock.app is well-organized with API reference, concept guides, SDK documentation, and integration guides. The workflow concept documentation explains triggers, steps, channels, conditions, batching, and digests. SDK documentation covers server (Node, Python, Ruby, Go, PHP, Java, Elixir, .NET) and client (JavaScript, React, iOS, Android) libraries. The in-app feed component documentation includes pre-built React components with customization guides. Channel provider integration guides cover SendGrid, Postmark, SES, APNs, FCM, Slack, and more. For agents, the workflow triggering guide and the concept documentation are essential starting points. The documentation effectively communicates the mental model shift from per-channel notification code to workflow-based orchestration.

Rhumb editorial team Mar 16, 2026

Use in your agent

mcp
get_score ("knock")
● Knock 7.1 L3 Ready
exec: 7.5 · access: 6.4

Trust & provenance

This score is documentation-derived. Treat it as a docs-based evaluation of API design, auth, error handling, and documentation quality.

Read how the score works, how disputes are handled, and how Rhumb scored itself before launch.

Overall tier

L3 Ready

7.1 / 10.0

Alternatives

No alternatives captured yet.