目录 / webhookrelay
MCP
鉴权未知
未评级
已上架
webhookrelay
该来源不提供完整文件导出(国内平台多为平台内托管),仅存元数据与原链
接入信息
- 传输形态
- http
- 鉴权方式
- 鉴权未知
- 端点
https://webhookrelay--karolis.run.tools
鉴权方式未标注,请核对官方文档后再接入——不要直接使用以下片段
{
"mcpServers": {
"webhookrelay": {
"url": "https://webhookrelay--karolis.run.tools"
}
}
}
能力清单
| 工具 | 说明 |
|---|---|
| attach_function | Attach a transform function to an input or output. |
| clear_output_rules | Remove all filtering rules from an output so it forwards every webhook again. |
| create_bucket | Create a new webhook bucket with a public endpoint. Optionally specify a destination URL to create an output in the same call. Returns the public URL where 3rd parties should send webhooks. Set internal=true when the destination is a local service (like localhost). Use input_function_id and/or output_function_id to attach transform functions during creation. |
| create_email_input | Create an inbound email-to-webhook input on a bucket. Returns a unique inbound email address ({token}@<inbound-domain>); mail sent to it is parsed into a structured payload and pushed through the bucket's transform + outputs, exactly like a webhook. No service connection or credentials needed. Optionally attach a transform function and restrict senders with allowed_senders. |
| create_function | Create a new transform function. Supported drivers: 'lua' for Lua scripts, 'js' for JavaScript. IMPORTANT: Before writing function code, you MUST read the 'webhookrelay://docs/functions/javascript-api' resource to understand the available runtime API. The runtime uses custom globals (r, cfg, http, crypto, time, bigquery, mailgun) that are NOT standard JavaScript/Node.js — do not use require(), import, fetch(), or browser APIs. For JavaScript transforms, mutate 'r' directly (for example r.setMethod(...), r.setBody(...)); do not define wrapper handlers like function handler(ctx). For more complicated functions, run execute with realistic payloads and iterate until behavior is correct. |
| create_input | Create an additional webhook input (public HTTPS endpoint) in a bucket. Each input gets its own endpoint URL where third parties send webhooks; all inputs in a bucket forward to the same outputs. By default the endpoint responds with HTTP 200 and an empty body; use response_status_code / response_body / response_headers to customize the static response, or response_from_output to respond with a destination's actual response. Note: create_bucket already creates a default input, so only use this when a bucket needs more than one endpoint. |
| create_output | Create a webhook output (forwarding destination) in a bucket. Every webhook received by the bucket's inputs is forwarded to each enabled output whose rules match. Use internal=true for destinations on private networks that are reached through a connected relay agent (localhost destinations are forced to internal automatically). For forwarding to cloud services like SQS or PubSub use create_service_connection_output instead. |
| create_service_connection | Create a new service connection to a cloud provider. Provide the service_type (gcp, aws, or azure) and the corresponding credentials. For GCP: gcp_project_id and gcp_service_account_key are required. For AWS: aws_access_key_id and aws_secret_access_key are required. For Azure: azure_client_id, azure_client_secret, and azure_tenant_id are required. |
| create_service_connection_input | Create a service connection input on a bucket. This subscribes to events from a cloud service and forwards them into the bucket. Requires a service_connection_id (for credentials) and an input_type. Supported types: gcp_pubsub (requires gcp_pubsub_subscription_name), gcp_gcs (requires gcp_gcs_bucket_name), aws_s3 (requires aws_s3_bucket_name and aws_s3_region), aws_sqs (requires aws_sqs_queue_url), aws_sns (requires aws_sns_topic_arn). |
| create_service_connection_output | Create a service connection output on a bucket. This forwards webhooks from the bucket to a cloud service. Requires a service_connection_id (for credentials) and an output_type. Supported types: gcp_pubsub (requires gcp_pubsub_topic_name), gcp_gcs (requires gcp_gcs_bucket_name), aws_s3 (requires aws_s3_bucket_name and aws_s3_region), aws_sqs (requires aws_sqs_queue_url), aws_sns (requires aws_sns_topic_arn). |
| delete_bucket | Delete a webhook bucket and all its inputs and outputs. |
| delete_input | Delete a webhook input. Its endpoint URL stops accepting webhooks immediately. Webhook logs are not affected. |
| delete_output | Delete a webhook output. Webhooks are no longer forwarded to its destination. If an input used this output for its response (response_from_output), that input reverts to its static response. |
| delete_service_connection | Delete a service connection by ID. |
| delete_service_connection_input | Delete a service connection input from a bucket. |
| delete_service_connection_output | Delete a service connection output from a bucket. |
| detach_function | Detach a transform function from an input or output. |
| execute | Execute a transform function with a synthetic request payload (equivalent to the invoke endpoint). For JavaScript runtime API details, read MCP resource: webhookrelay://docs/functions/javascript-api. For complex transforms, use execute iteratively with realistic payloads to debug and refine behavior. |
| get_function | Get full details of a transform function including its source code. |
| get_input | Get the full configuration of a webhook input (public HTTPS endpoint), including the response settings returned to webhook senders, custom domain, path prefix and TLS settings. Use list_buckets to discover input IDs. |
| get_logs_stats | Get webhook delivery statistics (received / success / failures counts). Two modes: by default returns one per-day series for the whole account (or a single bucket via bucket_id). Pass group_by=bucket to instead get a per-bucket leaderboard ranked by failures — the fastest way to find WHICH buckets are failing across the account in one call (then drill in with list_webhook_logs status=failed and retry with retry_webhook). group_by=bucket supports limit (default 10, max 100) and offset paging. Days are UTC; the from/to window is inclusive and defaults to the last 7 days. |
| get_output | Get the full configuration of a webhook output (forwarding destination), including override headers, retry/timeout settings, filtering rules and durable retry configuration. Use list_buckets to discover output IDs. |
| get_service_connection | Get details of a specific service connection by ID. Returns connection configuration with secrets redacted. |
| get_webhook_log | Get full details of a specific webhook log including request and response headers and body. Requires the bucket_id the log belongs to (returned by list_webhook_logs). |
| list_buckets | List all webhook buckets for the account. Each bucket contains inputs (public endpoints that receive webhooks) and outputs (destinations where webhooks are forwarded to). |
| list_functions | List all transform functions. Functions are code snippets (Lua or JavaScript) that can modify webhooks in transit. |
| list_service_connection_inputs | List all service connection inputs for a bucket. Service connection inputs subscribe to cloud events (e.g. GCP PubSub subscriptions, AWS SQS queues, AWS SNS topics, S3/GCS bucket notifications) and forward them into the bucket. |
| list_service_connection_outputs | List all service connection outputs for a bucket. Service connection outputs forward webhooks received by the bucket to cloud services (e.g. GCP PubSub topics, AWS SQS queues, AWS SNS topics, S3/GCS buckets). |
| list_service_connections | List all service connections (GCP, AWS, Azure) for the account. Service connections store cloud provider credentials used to subscribe to events (e.g. GCP PubSub, AWS SQS) or send webhooks to cloud services (e.g. S3, BigQuery). |
| list_webhook_logs | List webhook logs for a bucket in a lightweight format. Shows status code, status, method, timestamp and duration for each webhook event. Supports filtering by delivery status (e.g. status=failed to find broken deliveries), output ID and a from/to time window. Use get_webhook_log for full request/response details. |
| retry_webhook | Retry (resend) a webhook log through its bucket's delivery pipeline. The stored request is re-relayed to the bucket's outputs — or to a single output when output_id is set. process_policy controls whether rules and transform functions run again ('skip' or 'force'). Ephemeral logs cannot be retried. |
| set_output_rules | Set filtering rules on an output. The output only forwards webhooks for which the rule tree evaluates to true. The rule tree has exactly one top-level key: 'match' (a single condition), or 'and' / 'or' (arrays of nested rule trees), or 'not' (negates a nested tree). A 'match' condition has a 'type' and operands: {"match": {"type": "value", "value": "opened", "parameter": {"source": "payload", "name": "action"}}} matches when the JSON payload field 'action' equals 'opened'; types 'contains' / 'does-not-contain' use 'substring' instead of 'value'; type 'regex' uses 'regex'; types 'payload-hash-sha1' / 'payload-hash-sha256' verify an HMAC signature header using 'secret'; type 'ip-whitelist' uses 'ip-range' (CIDR). 'parameter.source' is one of: 'payload' (JSON body field, dot notation for nesting), 'header', 'query', 'path', 'string', 'entire-payload', 'entire-query', 'entire-headers'. Example - only forward GitHub push events from the main branch: {"and": [{"match": {"type": "value", "value": "refs/heads/main", "parameter": {"source": "payload", "name": "ref"}}}, {"match": {"type": "value", "value": "push", "parameter": {"source": "header", "name": "X-GitHub-Event"}}}]} |
| update_bucket | Update a webhook bucket's name or description. |
| update_function | Update an existing transform function's name, driver, or code. IMPORTANT: Before writing function code, you MUST read the 'webhookrelay://docs/functions/javascript-api' resource to understand the available runtime API. The runtime uses custom globals (r, cfg, http, crypto, time, bigquery, mailgun) that are NOT standard JavaScript/Node.js — do not use require(), import, fetch(), or browser APIs. For JavaScript transforms, mutate 'r' directly (for example r.setMethod(...), r.setBody(...)); do not define wrapper handlers like function handler(ctx). For more complicated functions, run execute with realistic payloads and iterate until behavior is correct. |
| update_input | Update a webhook input's configuration. Only the provided fields are changed; omitted fields keep their current values. Provide an empty string to clear a text field (for example function_id: "" detaches the transform function). |
| update_output | Update a webhook output's configuration. Only the provided fields are changed; omitted fields keep their current values. Provide an empty string to clear a text field (for example function_id: "" detaches the transform function). Filtering rules are managed separately with set_output_rules / clear_output_rules. |
| update_service_connection | Update a service connection's name or provider-specific credentials. Only non-empty fields are updated. |
纠错与举报(发现条目失效、署名有误或涉及侵权?)
提交举报 / 纠错
侵权举报经核验成立后,我们会即时下线该条目并删除已存的内容副本。