Feat(aigw): v2 log weather mcp #6525
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,373 @@ | ||
| --- | ||
| title: Monitor MCP traffic with the HTTP Log plugin | ||
| content_type: how_to | ||
| related_resources: | ||
| - text: "{{site.ai_gateway}}" | ||
| url: /ai-gateway/ | ||
| - text: AI MCP Server entity | ||
| url: /ai-gateway/entities/ai-mcp-server/ | ||
| - text: HTTP Log | ||
| url: /ai-gateway/policies/http-log/ | ||
|
|
||
| description: | | ||
| Create an AI MCP from a mock API using `conversion-listener` mode, then use the HTTP Log Policy to record and inspect the payloads and responses from the example MCP. | ||
|
|
||
| products: | ||
| - ai-gateway | ||
|
|
||
| permalink: /ai-gateway/monitor-mcp-traffic-with-http-log/ | ||
|
|
||
| series: | ||
| id: mcp-conversion-2-0 | ||
| position: 2 | ||
|
|
||
| works_on: | ||
| - konnect | ||
|
|
||
| min_version: | ||
| ai-gateway: '2.0' | ||
|
|
||
| tags: | ||
| - ai | ||
| - mcp | ||
|
|
||
| tldr: | ||
| q: How do I observe and validate MCP traffic from autogenerated tools? | ||
| a: | | ||
| Use an AI MCP entity to expose a mock API as an MCP tool, then use the HTTP Log Policy to capture tool calls and validate their payloads and responses. | ||
|
|
||
| tools: | ||
| - kongctl | ||
|
|
||
| cleanup: | ||
| inline: | ||
| - title: Destroy the {{site.base_gateway}} container | ||
| include_content: cleanup/products/gateway | ||
| icon_url: /assets/icons/gateway.svg | ||
| - title: Clean up {{site.konnect_short_name}} environment | ||
| include_content: cleanup/platform/konnect | ||
| icon_url: /assets/icons/gateway.svg | ||
|
|
||
| automated_tests: false | ||
|
|
||
| --- | ||
|
|
||
| ## Reconfigure the AI MCP Proxy plugin | ||
|
|
||
| To observe traffic for MCP tools, you first must **enable logging and statistics** on the AI MCP. Apply the below configuration to reconfigure the AI MCP to log full payloads: | ||
|
|
||
| ```sh | ||
| kongctl apply -f - --auto-approve --pat "$KONNECT_TOKEN" <<EOF | ||
| ai_gateway_mcp_servers: | ||
| - ref: marketplace-mcp | ||
| ai_gateway: !lookup {id: $AI_GATEWAY_ID} | ||
| name: marketplace-mcp | ||
| display_name: "Marketplace API" | ||
| type: conversion-listener | ||
| enabled: true | ||
| policies: [] | ||
| access: | ||
| acl_attribute_type: consumer | ||
| acls: | ||
| allow: [] | ||
| default_tool_acls: | ||
| deny: [] | ||
| config: | ||
| url: http://host.docker.internal:3000 | ||
| route: | ||
| paths: | ||
| - /marketplace | ||
| logging: | ||
| payloads: true | ||
| statistics: true | ||
| server: | ||
| timeout: 60000 | ||
| tools: | ||
| - name: get-users | ||
| description: Get users | ||
| method: GET | ||
| path: /marketplace/users | ||
| parameters: | ||
| - name: id | ||
| in: query | ||
| required: false | ||
| schema: | ||
| type: string | ||
| description: Optional user ID | ||
| - name: get-orders-for-user | ||
| description: Get orders for a user | ||
| method: GET | ||
| path: /marketplace/orders | ||
| parameters: | ||
| - description: User ID to filter orders | ||
| in: query | ||
| name: userid | ||
| required: true | ||
| schema: | ||
| type: string | ||
| EOF | ||
| ``` | ||
|
|
||
|
|
||
| ## Log MCP traffic | ||
|
|
||
| Before we send tool calls, we need to set up the a HTTP Logs Policy to check how many tokens we've managed to save by using our configuration. | ||
|
Check failure on line 114 in app/_how-tos/ai-gateway/monitor-mcp-traffic-with-http-log.md
|
||
|
|
||
| Apply the below config to create an HTTP logs Policy and attach it to the AI MCP entity: | ||
|
|
||
| ```sh | ||
| kongctl apply -f - --auto-approve --pat "$KONNECT_TOKEN" <<EOF | ||
| ai_gateway_policies: | ||
| - ref: my-http-log-policy | ||
| name: my-http-log-policy | ||
| ai_gateway: !lookup {id: $AI_GATEWAY_ID} | ||
| type: http-log | ||
| enabled: true | ||
| global: false | ||
| config: | ||
| http_endpoint: http://host.docker.internal:9999/ | ||
| headers: | ||
| Authorization: Bearer some-token | ||
| method: POST | ||
| timeout: 3000 | ||
|
|
||
| ai_gateway_mcp_servers: | ||
| - ref: marketplace-mcp | ||
| ai_gateway: !lookup {id: $AI_GATEWAY_ID} | ||
| name: marketplace-mcp | ||
| display_name: "Marketplace API" | ||
| type: conversion-listener | ||
| enabled: true | ||
| policies: [ !ref my-http-log-policy#name ] | ||
| access: | ||
| acl_attribute_type: consumer | ||
| acls: | ||
| allow: [] | ||
| default_tool_acls: | ||
| deny: [] | ||
| config: | ||
| url: http://host.docker.internal:3000 | ||
| route: | ||
| paths: | ||
| - /marketplace | ||
| logging: | ||
| payloads: true | ||
| statistics: true | ||
| server: | ||
| timeout: 60000 | ||
| tools: | ||
| - name: get-users | ||
| description: Get users | ||
| method: GET | ||
| path: /marketplace/users | ||
| parameters: | ||
| - name: id | ||
| in: query | ||
| required: false | ||
| schema: | ||
| type: string | ||
| description: Optional user ID | ||
| - name: get-orders-for-user | ||
| description: Get orders for a user | ||
| method: GET | ||
| path: /marketplace/orders | ||
| parameters: | ||
| - description: User ID to filter orders | ||
| in: query | ||
| name: userid | ||
| required: true | ||
| schema: | ||
| type: string | ||
| EOF | ||
| ``` | ||
|
|
||
| Let's run a simple log collector script which collect logs at `9999` port. Copy and run this snippet in your terminal: | ||
|
|
||
| ``` | ||
| cat <<EOF > log_server.py | ||
| from http.server import BaseHTTPRequestHandler, HTTPServer | ||
| import datetime | ||
|
|
||
| LOG_FILE = "kong_logs.txt" | ||
|
|
||
| class LogHandler(BaseHTTPRequestHandler): | ||
| def do_POST(self): | ||
| timestamp = datetime.datetime.now().isoformat() | ||
|
|
||
| content_length = int(self.headers['Content-Length']) | ||
| post_data = self.rfile.read(content_length).decode('utf-8') | ||
|
|
||
| log_entry = f"{timestamp} - {post_data}\n" | ||
| with open(LOG_FILE, "a") as f: | ||
| f.write(log_entry) | ||
|
|
||
| print("="*60) | ||
| print(f"Received POST request at {timestamp}") | ||
| print(f"Path: {self.path}") | ||
| print("Headers:") | ||
| for header, value in self.headers.items(): | ||
| print(f" {header}: {value}") | ||
| print("Body:") | ||
| print(post_data) | ||
| print("="*60) | ||
|
|
||
| # Send OK response | ||
| self.send_response(200) | ||
| self.end_headers() | ||
| self.wfile.write(b"OK") | ||
|
|
||
| if __name__ == '__main__': | ||
| server_address = ('', 9999) | ||
| httpd = HTTPServer(server_address, LogHandler) | ||
| print("Starting log server on http://0.0.0.0:9999") | ||
| httpd.serve_forever() | ||
| EOF | ||
| ``` | ||
|
|
||
| Now, run this script with Python and UV: | ||
|
|
||
| ```sh | ||
| uv run log_server.py | ||
| ``` | ||
|
|
||
| If the script is successful, you'll receive the following prompt in your terminal: | ||
|
|
||
| ```sh | ||
| Starting log server on http://0.0.0.0:9999 | ||
| ``` | ||
|
|
||
| ## Validate MCP tools configuration | ||
|
|
||
| You can validate that the HTTP Logs plugin is collecting metrics by generating MCP traffic to the mock API server. Enter the following question in Chatwise: | ||
|
|
||
| ```text | ||
| what has bob ordered ? | ||
| ``` | ||
|
|
||
| Once Chatwise has finished reasoning, you will see MCP audit log entries logged by the HTTP plugin in your terminal similar to the following: | ||
|
|
||
|
|
||
| ```sh | ||
| Received POST request at 2026-07-31T15:56:28.232507 | ||
| Path: / | ||
| Headers: | ||
| Content-Length: 5145 | ||
| Authorization: Bearer some-token | ||
| Content-Type: application/json | ||
| Host: host.docker.internal:9999 | ||
| User-Agent: lua-resty-http/0.17.2 (Lua) ngx_lua/10028 | ||
| Body: | ||
| { | ||
| "upstream_uri": "/", | ||
| "latencies": { | ||
| "redis": 0, | ||
| "socket": 0, | ||
| "request": 15, | ||
| "kong": 14, | ||
| "proxy": -1, | ||
| "third_party": 5.936128, | ||
| "client": 0.7616, | ||
| "receive": 0, | ||
| "http_client": 5.936128, | ||
| "kong_internal": 8.50176, | ||
| "dns": 0 | ||
| }, | ||
| "request": { | ||
| "headers": { | ||
| "mcp-protocol-version": "2025-11-25", | ||
| "content-length": "126", | ||
| "accept": "application/json, text/event-stream", | ||
| "connection": "keep-alive", | ||
| "accept-language": "en-GB", | ||
| "mcp-session-id": "8b2d1534-3ef9-4708-9baf-5cf36b4f5b1f", | ||
| "content-type": "application/json", | ||
| "host": "localhost:8000", | ||
| "user-agent": "ChatWise/26.7.4" | ||
| }, | ||
| "size": 483, | ||
| "method": "POST", | ||
| "url": "http://localhost:8000/marketplace", | ||
| "id": "cc054f73a6b04f3573cdf02883d92f53", | ||
| "uri": "/marketplace", | ||
| "querystring": {} | ||
| }, | ||
| "route": { | ||
| "service": { | ||
| "id": "c16f08e6-4cfe-5c39-8627-f744f091e592" | ||
| }, | ||
| "regex_priority": 0, | ||
| "preserve_host": false, | ||
| "protocols": [ | ||
| "http", | ||
| "https" | ||
| ], | ||
| "path_handling": "v0", | ||
| "name": "marketplace-mcp-route", | ||
| "id": "a878697a-8bbc-591b-8640-e1bd5aba261c", | ||
| "ws_id": "17dadd2f-94c5-4ba4-86bb-8d2d4e6e22cd", | ||
| "paths": [ | ||
| "/marketplace" | ||
| ], | ||
| "updated_at": 1785509686, | ||
| "strip_path": true, | ||
| "https_redirect_status_code": 426, | ||
| "response_buffering": true, | ||
| "request_buffering": true, | ||
| "created_at": 1785509686 | ||
| }, | ||
| "started_at": 1785509788214, | ||
| "client_ip": "192.168.165.1", | ||
| "workspace": "17dadd2f-94c5-4ba4-86bb-8d2d4e6e22cd", | ||
| "workspace_name": "default", | ||
| "ai": { | ||
| "mcp": { | ||
| "mcp_session_id": "8b2d1534-3ef9-4708-9baf-5cf36b4f5b1f", | ||
| "rpc": [ | ||
| { | ||
| "id": "2", | ||
| "method": "tools/call", | ||
| "response_body_size": 2175, | ||
| "tool_name": "get-orders-for-user", | ||
| "latency": 15, | ||
| "payload": { | ||
| "request": "{\"id\":2,\"params\":{\"name\":\"get-orders-for-user\",\"arguments\":{\"query_userid\":\"e5f6g7h8\"}},\"jsonrpc\":\"2.0\",\"method\":\"tools/call\"}", | ||
| "response": "{\"id\":2,\"jsonrpc\":\"2.0\",\"result\":{\"content\":[{\"type\":\"text\",\"text\":\"[{\\\"id\\\":\\\"ord001\\\",\\\"name\\\":\\\"Sugar (50kg)\\\",\\\"userId\\\":\\\"a1b2c3d4\\\"},{\\\"id\\\":\\\"ord002\\\",\\\"name\\\":\\\"Cleaning Supplies Pack\\\",\\\"userId\\\":\\\"a1b2c3d4\\\"},{\\\"id\\\":\\\"ord003\\\",\\\"name\\\":\\\"Canned Tomatoes (100 cans)\\\",\\\"userId\\\":\\\"a1b2c3d4\\\"},{\\\"id\\\":\\\"ord004\\\",\\\"name\\\":\\\"Flour (100kg)\\\",\\\"userId\\\":\\\"e5f6g7h8\\\"},{\\\"id\\\":\\\"ord005\\\",\\\"name\\\":\\\"Dish Soap (10 bottles)\\\",\\\"userId\\\":\\\"e5f6g7h8\\\"},{\\\"id\\\":\\\"ord006\\\",\\\"name\\\":\\\"Salt (25kg)\\\",\\\"userId\\\":\\\"e5f6g7h8\\\"},{\\\"id\\\":\\\"ord007\\\",\\\"name\\\":\\\"Olive Oil (20L)\\\",\\\"userId\\\":\\\"i9j0k1l2\\\"},{\\\"id\\\":\\\"ord008\\\",\\\"name\\\":\\\"Baking Powder (10kg)\\\",\\\"userId\\\":\\\"i9j0k1l2\\\"},{\\\"id\\\":\\\"ord009\\\",\\\"name\\\":\\\"Rice (200kg)\\\",\\\"userId\\\":\\\"m3n4o5p6\\\"},{\\\"id\\\":\\\"ord010\\\",\\\"name\\\":\\\"Vegetable Oil (15L)\\\",\\\"userId\\\":\\\"m3n4o5p6\\\"},{\\\"id\\\":\\\"ord011\\\",\\\"name\\\":\\\"Pasta (80kg)\\\",\\\"userId\\\":\\\"m3n4o5p6\\\"},{\\\"id\\\":\\\"ord012\\\",\\\"name\\\":\\\"Canned Beans (50 cans)\\\",\\\"userId\\\":\\\"m3n4o5p6\\\"},{\\\"id\\\":\\\"ord013\\\",\\\"name\\\":\\\"Toilet Paper (Case of 48)\\\",\\\"userId\\\":\\\"q7r8s9t0\\\"},{\\\"id\\\":\\\"ord014\\\",\\\"name\\\":\\\"Hand Sanitizer (20 bottles)\\\",\\\"userId\\\":\\\"q7r8s9t0\\\"},{\\\"id\\\":\\\"ord015\\\",\\\"name\\\":\\\"Laundry Detergent (10L)\\\",\\\"userId\\\":\\\"u1v2w3x4\\\"},{\\\"id\\\":\\\"ord016\\\",\\\"name\\\":\\\"Trash Bags (100 ct)\\\",\\\"userId\\\":\\\"u1v2w3x4\\\"},{\\\"id\\\":\\\"ord017\\\",\\\"name\\\":\\\"Disinfectant Spray (5 bottles)\\\",\\\"userId\\\":\\\"u1v2w3x4\\\"},{\\\"id\\\":\\\"ord018\\\",\\\"name\\\":\\\"Coffee Beans (30kg)\\\",\\\"userId\\\":\\\"k7l8m9n0\\\"},{\\\"id\\\":\\\"ord019\\\",\\\"name\\\":\\\"Tea Bags (500ct)\\\",\\\"userId\\\":\\\"k7l8m9n0\\\"},{\\\"id\\\":\\\"ord020\\\",\\\"name\\\":\\\"Condensed Milk (40 cans)\\\",\\\"userId\\\":\\\"k7l8m9n0\\\"},{\\\"id\\\":\\\"ord021\\\",\\\"name\\\":\\\"Paper Towels (24 rolls)\\\",\\\"userId\\\":\\\"g3h4i5j6\\\"},{\\\"id\\\":\\\"ord022\\\",\\\"name\\\":\\\"Broom & Mop Set\\\",\\\"userId\\\":\\\"g3h4i5j6\\\"},{\\\"id\\\":\\\"ord023\\\",\\\"name\\\":\\\"Cereal (20 boxes)\\\",\\\"userId\\\":\\\"c9d0e1f2\\\"},{\\\"id\\\":\\\"ord024\\\",\\\"name\\\":\\\"Powdered Milk (10kg)\\\",\\\"userId\\\":\\\"c9d0e1f2\\\"},{\\\"id\\\":\\\"ord025\\\",\\\"name\\\":\\\"Snacks Variety Pack\\\",\\\"userId\\\":\\\"c9d0e1f2\\\"},{\\\"id\\\":\\\"ord026\\\",\\\"name\\\":\\\"Cooking Gas Cylinder\\\",\\\"userId\\\":\\\"y5z6a7b8\\\"},{\\\"id\\\":\\\"ord027\\\",\\\"name\\\":\\\"Napkins (1000ct)\\\",\\\"userId\\\":\\\"y5z6a7b8\\\"}]\"}],\"isError\":false}}" | ||
| } | ||
| } | ||
| ] | ||
| } | ||
| }, | ||
| "tries": [], | ||
| "response": { | ||
| "status": 200, | ||
| "headers": { | ||
| "connection": "close", | ||
| "cache-control": "no-store, no-transform", | ||
| "transfer-encoding": "chunked", | ||
| "x-kong-request-id": "cc054f73a6b04f3573cdf02883d92f53", | ||
| "server": "kong/2.0.1-ai-gateway", | ||
| "content-type": "text/event-stream", | ||
| "x-kong-response-latency": "14" | ||
| }, | ||
| "size": 2503 | ||
| }, | ||
| "service": { | ||
| "name": "marketplace-mcp", | ||
| "ws_id": "17dadd2f-94c5-4ba4-86bb-8d2d4e6e22cd", | ||
| "host": "host.docker.internal", | ||
| "write_timeout": 60000, | ||
| "connect_timeout": 60000, | ||
| "updated_at": 1785509686, | ||
| "read_timeout": 60000, | ||
| "retries": 5, | ||
| "id": "c16f08e6-4cfe-5c39-8627-f744f091e592", | ||
| "created_at": 1785509686, | ||
| "port": 3000, | ||
| "enabled": true, | ||
| "protocol": "http" | ||
| }, | ||
| "source": "kong", | ||
| "upstream_status": "" | ||
| } | ||
|
|
||
| ``` | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@jbaross - can we just use the file log policy instead of this script? It makes this flow much simpler
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think that would be an entirely separate how-to, using the http-log is the point of this one