Capture and inspect HTTP requests, emails, and DNS lookups to test webhooks, debug API callbacks, and perform security testing.
Also known as zebbern/webhook-mcp-server, webhook-mcp-server
- Stars
- 12
- Tools
- 32
- Downloads, 30 days
- 243
- Latest version
- 3.0.0
- Last commit
- Sep 3, 2026
- License
- MIT
What is Webhook.site MCP?
Webhook.site is an MCP server. Capture and inspect HTTP requests, emails, and DNS lookups to test webhooks, debug API callbacks, and perform security testing. It exposes 32 tools, such as server_status, create_webhook, configure_webhook.
Summary generated from the project's README
About Webhook.site MCP server
Developers and security researchers can capture and inspect HTTP requests, emails, and DNS lookups through this MCP server. It is designed for testing webhooks, debugging API callbacks, and performing authorized security testing like bug bounty hunting. By connecting to an AI client such as Claude Desktop, Cursor, or VS Code, users can interact with the webhook.site service to monitor real-time data or simulate specific API responses.
Users can operate the server anonymously with 7-day URLs or provide a WEBHOOK_SITE_API_KEY to access permanent URLs and advanced account features. The server includes 31 tools, such as create_webhook for generating disposable inboxes, wait_for_request to pause until a callback arrives, and generate_oob_payloads for creating SSRF or XSS testing payloads. It also supports complex automation through Custom Actions, Schedules, and SQL-based Databases when an API key is configured.
What you can do with it
- 1Debug API callbacksA developer can create a webhook and wait for a specific HTTP request to arrive to inspect its headers and body.
- 2Test sign-up flowsAn AI client can generate a temporary email, wait for a password reset link, and then follow that link to verify a user account.
- 3Perform security testingSecurity researchers can generate out-of-band payloads to detect blind vulnerabilities like SSRF or XSS.
- 4Monitor DNS activityUsers can set up a DNS hook to detect if a server is making unauthorized DNS queries.
Written from the project's README. Every use case is checked against a quote from it.
How to install Webhook.site MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall Webhook.site MCP in Claude Code
$ claude mcp add webhook-mcp-server -- uvx webhook-mcp-serverInstall Webhook.site MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"webhook-mcp-server": {
"command": "uvx",
"args": [
"webhook-mcp-server"
]
}
}
}Install Webhook.site MCP in VS Code
.vscode/mcp.json
{
"servers": {
"webhook-mcp-server": {
"type": "stdio",
"command": "uvx",
"args": [
"webhook-mcp-server"
]
}
}
}Install Webhook.site MCP in VS Code CLI
$ code --add-mcp '{"name":"webhook-mcp-server","type":"stdio","command":"uvx","args":["webhook-mcp-server"]}'Install Webhook.site MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"webhook-mcp-server": {
"command": "uvx",
"args": [
"webhook-mcp-server"
]
}
}
}Install Webhook.site MCP in Codex
$ codex mcp add webhook-mcp-server -- uvx webhook-mcp-serverInstall Webhook.site MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.webhook-mcp-server] command = "uvx" args = ["webhook-mcp-server"]
Install Webhook.site MCP in Gemini CLI
$ gemini mcp add -s user webhook-mcp-server uvx -- webhook-mcp-serverInstall Webhook.site MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"webhook-mcp-server": {
"command": "uvx",
"args": [
"webhook-mcp-server"
]
}
}
}Webhook.site MCP tools 32
Detected from the READMEserver_statuscreate_webhookconfigure_webhookget_webhook_infoget_webhook_emaillist_webhooksdelete_webhooksend_requestsget_webhook_requestssearch_requestsget_requestupdate_request
| server_status | — | |
| create_webhook | — | |
| configure_webhook | — | |
| get_webhook_info | — | |
| get_webhook_email | — | |
| list_webhooks | — | |
| delete_webhook | — | |
| send_requests | — | |
| get_webhook_requests | — | |
| search_requests | — | |
| get_request | — | |
| update_request | — |
Show all 32 toolsShow fewer tools
download_request_filedelete_requestdelete_all_requestsexport_webhook_datawait_for_requestwait_for_emailrespond_to_next_requestfollow_email_linkmanage_custom_actionsmanage_schedulesmanage_global_variablesmanage_groupsmanage_queuesmanage_templatesmanage_databasesmanage_usersgenerate_oob_payloadscheck_for_callbacksextract_links_from_requestget_latest_request
| download_request_file | — | |
| delete_request | — | |
| delete_all_requests | — | |
| export_webhook_data | — | |
| wait_for_request | — | |
| wait_for_email | — | |
| respond_to_next_request | — | |
| follow_email_link | — | |
| manage_custom_actions | — | |
| manage_schedules | — | |
| manage_global_variables | — | |
| manage_groups | — | |
| manage_queues | — | |
| manage_templates | — | |
| manage_databases | — | |
| manage_users | — | |
| generate_oob_payloads | — | |
| check_for_callbacks | — | |
| extract_links_from_request | — | |
| get_latest_request | — |
Automated checks
Ran Nov 13, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachablezebbern/webhook-mcp-server is public and reachablegithub.com/zebbern/webhook-mcp-server ↗Pass
- Package names this repository as its sourcewebhook-mcp-server names zebbern/webhook-mcp-server as sourcepypi.org/project/webhook-mcp-server/ ↗Pass
- Build provenance (Sigstore attestation)Sigstore attestation for webhook-mcp-serverpypi.org/project/webhook-mcp-server/ ↗Pass
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-09-03github.com/zebbern/webhook-mcp-server ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from webhook-mcp-serverPass
- No critical advisory in the package or its dependencies44 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model32 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubWebhook.site MCP Server
A Model Context Protocol (MCP) server for webhook.site - instantly capture HTTP requests, emails, and DNS lookups. Perfect for testing webhooks, debugging API callbacks, security testing, and bug bounty hunting.
Security helper tools (SSRF, XSS, canary tokens) are for authorized testing only — systems you own or have explicit permission to test.
Table of Contents
- Quick Start
- What Can You Do?
- Tools Reference
- Examples
- Each Webhook Token Provides
- Documentation
- Development
- Requirements
- Contributing
- Credits
- Links
Quick Start
Installation
# Using uvx (recommended - no install needed)
uvx webhook-mcp-server==3.0.0
# Or install via pip
pip install webhook-mcp-server==3.0.0
Use 3.0.0 or newer. 2.1.3 does not start on MCP 2.0; 3.0 renames a few tools (see Upgrading from 2.x).
VS Code / GitHub Copilot
Add to .vscode/mcp.json:
{
"servers": {
"webhook-mcp-server": {
"type": "stdio",
"command": "uvx",
"args": ["webhook-mcp-server==3.0.0"]
}
}
}
Cursor
Add to .cursor/mcp.json (project) or your user MCP config:
{
"mcpServers": {
"webhook-mcp-server": {
"command": "uvx",
"args": ["webhook-mcp-server==3.0.0"]
}
}
}
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"webhook-mcp-server": {
"command": "uvx",
"args": ["webhook-mcp-server==3.0.0"]
}
}
}
Configuration
All settings are environment variables on the server process. Put them in the env block of your MCP client config.
| Variable | Purpose |
|---|---|
WEBHOOK_SITE_API_KEY |
Your webhook.site API key. Makes new URLs permanent (your plan's quota) and unlocks list_webhooks, Custom Actions, Schedules, Global Variables, Groups, Queue Profiles, Templates, Databases, Users and CSV export |
WEBHOOK_SITE_DEFAULT_EXPIRY |
Seconds until new URLs expire when the call does not pass expiry. Unset means permanent on a paid account (7 days for anonymous URLs) |
WEBHOOK_MCP_RATE_LIMIT_MAX_WAIT |
When webhook.site answers 429 with a Retry-After up to this many seconds (default 15), the call waits and retries once; longer waits are reported as an error naming the wait |
FOLLOW_EMAIL_LINK_ALLOW_HOSTS |
follow_email_link only opens links to public internet hosts, and connects to the address it checked (no DNS rebinding). To test a sign-up flow on your own machine or intranet, list what to allow: localhost,127.0.0.1,*.corp.example,10.0.0.0/8 |
HTTPS_PROXY / HTTP_PROXY / NO_PROXY |
Honoured by follow_email_link. The proxy makes the connection, so the target is checked before the request instead of being pinned |
SSL_CERT_FILE / SSL_CERT_DIR |
Custom CA bundle for follow_email_link, for corporate TLS interception |
If the request to a verification link completes and then redirects somewhere that is not allowed (a local dev server, an intranet dashboard), the tool reports that hop's status with the refused target in blocked_redirect instead of failing: the request that consumed the token already went through, whether or not the site counted it as verified.
What Can You Do?
Capture Webhooks
"Create a webhook and show me the URL"
"What requests have been sent to my webhook?"
"Wait for a request to come in"
Security/Bug Bounty:
"Generate an SSRF payload to test for blind vulnerabilities"
"Create XSS callback payloads to detect blind XSS attacks"
"Make me a canary token to detect if someone accesses a URL"
Email Automation:
"Create a temp email and wait for a password reset link"
"Monitor this webhook for emails and extract all links from them"
"Give me 3 temporary emails at once" (batch creation)
API Testing:
"Create a webhook that returns a 404 error with a custom message"
"Make a webhook with CORS enabled that waits 5 seconds before responding"
"Send 10 different test requests to a webhook and show me all the captured data"
"Hold the next request and answer it with a 402 and this JSON body"
Real-time Monitoring:
"Create a webhook and wait for any HTTP request to arrive"
"Monitor for DNS lookups to detect if a server is making DNS queries"
"Search all requests for ones containing 'password' in the body"
Data Analysis:
"Export all captured webhook requests to JSON format"
"Show me statistics on requests received in the last hour"
"Filter and show only POST requests with specific headers"
Automate with your account:
"Add a Custom Action that forwards every request to my staging API"
"Schedule a health check of https://example.com every 5 minutes and alert on a bad status"
"Write a WebhookScript that answers with the request's JSON field 'id'"
Canary Tokens
"Create a canary URL to track document access"
"Generate a DNS canary for the config file"
"Set up an email tracker pixel"
Tools Reference
Everything works without an account on anonymous 7-day URLs. With WEBHOOK_SITE_API_KEY set, URLs are permanent and the account tools unlock the features of your plan. Every webhook_token accepts the UUID, an alias, a pasted https://webhook.site/... URL or the inbox address. Full parameters for each tool: docs/TOOLS.md.
31 tools, generated from the server by scripts/gen_tool_docs.py.
Diagnostics
| Tool | What it does |
|---|---|
server_status |
Check this server's setup: API key, account reachability, plan, real-time socket, env config. |
Webhooks
| Tool | What it does |
|---|---|
create_webhook |
Create a disposable inbox to sign up on a website: HTTP URL, temp email, DNS. |
configure_webhook |
Create a webhook with custom settings, or update one (pass webhook_token). |
get_webhook_info |
Show a webhook's settings, expiry, request count and every address. |
get_webhook_email |
Return the temp inbox to sign up, verify, magic-link, or reset a password. |
list_webhooks |
List the webhooks (URLs / inboxes) in the account. Needs WEBHOOK_SITE_API_KEY. |
delete_webhook |
Permanently delete a webhook and every captured request/email. |
Requests
| Tool | What it does |
|---|---|
send_requests |
Send one JSON body (data) or several (payloads) to the webhook URL to test capture. |
get_webhook_requests |
List captured HTTP, email, or DNS events for a webhook, one page at a time. |
search_requests |
Search captured events by method, body text, headers, type, or date. |
get_request |
Return one captured event: the newest by default, or request_id. |
update_request |
Attach a note to a captured request, or set its dynamic response. |
download_request_file |
Download an uploaded file or email attachment (base64) by its file_id. |
delete_request |
Delete one captured HTTP, email, or DNS event by request id. |
delete_all_requests |
Clear captured events on a webhook, optionally by date or search query. |
export_webhook_data |
Full dump of captured events with HTML and untruncated bodies, as JSON or CSV. |
Real-time
| Tool | What it does |
|---|---|
wait_for_request |
Wait until a new HTTP (or DNS) callback hits the webhook (1-120s). |
wait_for_email |
Wait for a sign-up, verify, magic-link, or password-reset email (1-120s). |
respond_to_next_request |
Hold the next request that hits the webhook and answer it with your own status, headers and body. |
follow_email_link |
Open the verify / magic-link / reset URL from a captured sign-up email. |
Account features (API key)
| Tool | What it does |
|---|---|
manage_custom_actions |
Manage the Custom Actions webhook.site runs on every request or email a token receives. |
manage_schedules |
Manage Schedules: webhook.site calls request_url on an interval (needs API key). |
manage_global_variables |
Manage Global Variables shared by all URLs, usable as $name$ in Custom Actions and Schedules. |
manage_groups |
Manage Groups that organise the account's webhooks (needs API key). |
manage_queues |
Manage Queue Profiles that throttle queued Custom Actions (needs API key). |
manage_templates |
Manage Templates: reusable sets of Custom Actions plus predefined variables (needs API key). |
manage_databases |
Manage webhook.site Databases and run SQL against them (needs API key). |
manage_users |
Manage team users on an Enterprise account (needs an administrator API key). |
Security testing
| Tool | What it does |
|---|---|
generate_oob_payloads |
Build authorized out-of-band payloads that ping this webhook: SSRF URLs, XSS callbacks, or canary tokens. |
check_for_callbacks |
See if SSRF, XSS, or canary callbacks arrived in the last N minutes. |
extract_links_from_request |
Pull confirm, reset, magic-link, and other URLs from a captured email or HTTP body. |
Upgrading from 2.x
| 2.x tool | 3.0 replacement |
|---|---|
create_webhook_with_config, update_webhook |
configure_webhook (pass webhook_token to update) |
get_webhook_url, get_webhook_dns |
fields of get_webhook_info |
get_latest_request |
get_request |
send_to_webhook, send_multiple_requests |
send_requests (data or payloads) |
generate_ssrf_payload, generate_xss_callback, generate_canary_token |
generate_oob_payloads (kind=ssrf|xss|canary, canary_type) |
Examples
Sign up on a website
create_webhook— getemail({token}@emailhook.site)- Use that address on the site (sign-up, verify, magic link, or password reset)
wait_for_email— receive the message, confirm / login / reset URLs, and any OTPfollow_email_linkto open a verify link, or typeverification_codeson the site
If you already have a token, get_webhook_email returns the same inbox.
Create a Webhook
// Response from create_webhook
{
"token": "abc123-def456-...",
"url": "https://webhook.site/abc123-def456-...",
"email": "abc123-def456-...@emailhook.site",
"dns": "abc123-def456-....dnshook.site"
}
Wait for Password Reset Email
// Response from wait_for_email
{
"email_received": true,
"subject": "Password Reset Request",
"from": "noreply@example.com",
"auth_links": ["https://example.com/reset?token=xyz789"],
"verification_codes": ["847291"]
}
Poll for new events without paging
// get_webhook_requests(webhook_token, since=<next_since from the previous call>)
{
"requests": [ ... only what arrived after the cursor ... ],
"next_since": 1788472548944304
}
SSRF Testing Payload
// Response from generate_oob_payloads(kind="ssrf")
{
"payloads": {
"http": "https://webhook.site/token?id=ssrf-test",
"dns": "ssrf-test.token.dnshook.site",
"ip_decimal": "http://2130706433/token",
"ip_hex": "http://0x7f000001/token"
}
}
Each Webhook Token Provides
| Endpoint | Format | Use Case |
|---|---|---|
| HTTP URL | https://webhook.site/{token} |
Capture HTTP/HTTPS requests (any sub-path too) |
| Subdomain | https://{token}.webhook.site |
Alternative URL format |
| Forced status | https://webhook.site/{token}/{status} |
Answer with that status, for retry-logic tests |
{token}@emailhook.site |
Capture incoming emails | |
| DNS | {token}.dnshook.site |
Capture DNS lookups (every subdomain) |
Documentation
| Page | Contents |
|---|---|
| docs/TOOLS.md | Every tool with parameters and hints, generated from the server |
| docs/webhook-site-notes.md | What webhook.site actually does, verified live, including where the official docs are wrong |
| docs/testing.md | The four test layers, recordings, verification scripts, model evals, account safety |
| docs/releasing.md | Version bumps, tagging, the publish workflow |
| CHANGELOG.md | Version history |
Development
git clone https://github.com/zebbern/webhook-mcp-server.git
cd webhook-mcp-server
pip install -e ".[dev]"
pytest -m "not live and not live_auth" # offline suite incl. replay of recorded live exchanges
python server.py # run the server on stdio
The offline suite replays real recorded webhook.site exchanges; the live tool check and the live / live_auth tiers hit the real API. Details in docs/testing.md.
Layout: server.py (entry point and lifespan), handlers/tools.py (tool registrations), services/ (one module per API area), utils/ (HTTP client, real-time socket, URL safety, references), scripts/ (live check, generators, verifiers), evals/ (model-in-the-loop prompts).
Requirements
- Python 3.10+
mcp >= 2.0.0httpx >= 0.26.0,httpcore >= 1.0.0,anyio >= 4.0.0python-socketio[asyncio_client] >= 5.11.0(real-time waiting; the tools fall back to polling without it)
Contributing
Bug reports and PRs are welcome. Keep a PR to one change, add a test whose expectation was checked against the real API (see docs/testing.md), and run python scripts/gen_tool_docs.py after touching a tool.
Credits
This project is not affiliated with or endorsed by webhook.site
Links
- 📦 PyPI Package
- 🐙 GitHub Repository
- 🌐 webhook.site - The service this MCP wraps
- 📖 Model Context Protocol - MCP specification
Made with ❤️ for the MCP community
Webhook.site MCP questions
How do I add Webhook.site MCP to Claude Code?
Run claude mcp add webhook-mcp-server -- uvx webhook-mcp-server in your terminal. Add --scope user to make it available in every project.
How do I add Webhook.site MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
What tools does Webhook.site MCP provide?
It exposes 32 tools, including server_status, create_webhook, configure_webhook, get_webhook_info, get_webhook_email.
How can I use this for security testing?
You can generate SSRF payloads, XSS callback payloads, and canary tokens to detect blind vulnerabilities or unauthorized access.
Do I need an account to use the server?
No, everything works without an account using anonymous 7-day URLs, but an API key unlocks permanent URLs and advanced account features.