@langwatch/mcp-server 0.0.3 → 0.0.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -1,18 +1,20 @@
1
1
  # LangWatch 🏰 MCP Server
2
2
 
3
- The LangWatch MCP Server is a tool designed to aid finding, searching, and looking up Traces from the LangWatch platform via the [Model Context Protocol](https://modelcontextprotocol.io/introduction).
3
+ The LangWatch MCP Server is a tool designed to aid finding, searching, and looking up LLM traces from the LangWatch platform via the [Model Context Protocol](https://modelcontextprotocol.io/introduction).
4
4
 
5
- This server facilitates communication with LangWatch's API, allowing you to retrieve and manage LLM traces.
5
+ This server facilitates LLM development by allowing the agent to search for traces, understand all the steps in between a problematic output and try to fix the issue.
6
+
7
+ ## Setup in your Codebase
8
+
9
+ Check out [LangWatch integration guide](https://docs.langwatch.ai/integration/overview) to start tracking your agents so both you and Cursor/Windsurf/Claude Code or your favorite coding assistant can debug it.
6
10
 
7
11
  ## Setup in Cursor 👩‍💻
8
12
 
9
13
  1. Navigate to the Cursor Settings
10
14
  2. Navigate to the MCP item in the sidebar
11
- 3. Set the "name" as "LangWatch MCP"
15
+ 3. Set the "name" as "LangWatch"
12
16
  4. Set the "type" to `command`
13
17
  5. Set the "command" to `npx -y @langwatch/mcp-server --apiKey=sk-lw-...`
14
- - You **must** use `=` to join the command with the value!
15
-
16
18
  - `--apiKey`: Your LangWatch API key. This is mandatory and must be provided.
17
19
  - `--endpoint`: *Optional* The endpoint for the LangWatch API. Defaults to `https://app.langwatch.ai` if not specified.
18
20
 
@@ -54,7 +56,7 @@ To use these tools within Cursor, follow these steps:
54
56
  3. **Interact with your Agent:**
55
57
  - Ask a question like the following to test the tools are accessible: *Note: When the tool is detected, you'll need to run `Run tool` in the chat view for it to be called.
56
58
 
57
- > "I just ran into an issue while debugging, show me the latest traces from the last 30 days"
59
+ > "I just ran into an issue while debugging, can you check the latest traces and fix it?"
58
60
 
59
61
  <picture>
60
62
  <source media="(prefers-color-scheme: dark)" srcset="../assets/mcp-server/cursor-example.dark.webp">