New — webhooks your AI agents can wait on. Hook & Sense opening to early access.

Use case · Scaling a Team

When your team outgrows email alerts.

Solo devs run on email. Growing teams run on Slack and Discord. The moment a second person needs to see a callback in real time, the inbox-and-screenshare workflow breaks — and that's exactly the gap Sense fills.

What's painful today

  • Callback failures land in one developer's inbox; the rest of the team finds out hours later — at standup, or when a customer complains.
  • Three people sharing the same Hook-plan login means audit logs lie about who replayed what, and seat limits cap real collaboration at the wrong size.
  • Stripe lives in one channel, GitHub in another, Shopify in a third — but no tool mirrors the captured callback into the channel where the team already is.

How HookSense solves it

HookSense is the callback layer for AI agents: an agent calls create_callback_endpoint, awaits the result with wait_for_callback, and trusts it because verify_signature already checked it.

Slack / Discord mirroring of every callback

Forward every captured callback into a Slack or Discord channel — method, status, provider, HMAC badge, and a link back to the dashboard. Your team sees the live stream where they already work, not in a separate dashboard.

10 team seats, real role separation

Sense lifts Hook's 3-seat cap to 10 — enough for a backend team plus QA, an SRE-on-call, and a PM who just wants to watch. Owner / admin / member / viewer roles map to who can edit, who can replay, and who can only look.

Unlimited endpoints + 90-day retention

Lifts Hook's 15-endpoint cap entirely. Teams running 8+ provider integrations (Stripe + Stripe Connect + GitHub + Shopify + Slack + Twilio + SendGrid + custom) finally fit, and incident post-mortems can look back three months.

See it in code

Wire a Slack channel to receive every captured webhook

# 1. In Slack: Apps → Incoming Webhooks → Add to Slack → pick channel
# 2. Copy the webhook URL (https://hooks.slack.com/services/...)
# 3. In HookSense: endpoint → Forwards panel → Add forward

# Or via API:
curl -X POST https://hooksense.com/api/endpoints/<slug>/forwards \
  -H "Authorization: Bearer hsk_..." \
  -H "Content-Type: application/json" \
  -d '{
    "type": "slack",
    "targetUrl": "https://hooks.slack.com/services/T.../B.../...",
    "includeBody": true
  }'

# Every webhook captured at /w/<slug> now also posts to your Slack channel.
# Max 10 forwards per endpoint. Fire-and-forget — never blocks capture.

FAQ

How is Forwarding different from Alerts?

Forwarding mirrors every callback into a channel — including the successful ones — so the team has live visibility into the stream. Alerts only fire when a rule trips (failure rate, inactivity, latency spike). Most teams run both: a #stripe-stream channel for the firehose, a #stripe-incidents channel for paging.

What else comes with Sense beyond team features?

Sense is the production tier: alongside Slack/Discord forwarding and 10 seats, it adds a health dashboard, alerts on failures and inactivity, and custom domains. Automatic retries with exponential backoff for failed callbacks are coming soon. Paid plans are currently early-access.

Will Slack/Discord get spammed during a load test?

Each forward has an enabled toggle — mute it before a load test, re-enable when you're done. Body previews are also capped at 1,500 characters, so even chatty events don't blow up the channel.

What if our team already pages from PagerDuty / Opsgenie?

Use Alerts (with email or Slack to a PD-integrated channel) for incident routing, and Forwards for visibility into the normal stream. PagerDuty integrations sit downstream of Slack alerts in most setups — no need for HookSense to integrate with them directly.

Give your agent a callback URL in 30 seconds.

Run npx @hooksense/mcp in Claude Desktop, Cursor, or any MCP client. Free plan, no signup required. Paid plans are early-access.

Create a callback endpoint

Other use cases