molecule-ai/docs
35
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs(mcp-server): add LOG_LEVEL env var + structured pino logging section (#78)

Browse Source 
Pair PR: molecule-mcp-server#6
- Adds LOG_LEVEL to Environment Variables table (pino log levels: trace→fatal)
- Adds Structured Logging section explaining AsyncLocalStorage context,
  JSON output in production, pretty-print in dev
- Adds troubleshooting row for high-volume log output
- Links to molecule-mcp-server#6 for implementation details

Co-authored-by: Molecule AI Documentation Specialist <documentation-specialist@agents.moleculesai.app>
This commit is contained in:
molecule-ai[bot] 2026-04-22 23:57:30 +00:00 committed by GitHub
parent 40bd0cfdde
commit 4223530f35
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 27 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
content/docs/guides/mcp-server-setup.md
View File

@ -53,6 +53,32 @@ list_workspaces
| Variable | Default | Description |
|----------|---------|-------------|
| `MOLECULE_URL` | `http://localhost:8080` | Platform API URL |
| `LOG_LEVEL` | `info` | Log level — `trace`, `debug`, `info`, `warn`, `error`, `fatal` (pino numeric levels: 10–60) |
## Structured Logging
The MCP server emits structured JSON logs via [pino](https://github.com/pinojs/pino). Each log entry includes `level`, `time`, `pid`, `hostname`, `msg`, and any extra fields passed as context.
When `NODE_ENV != "production"`, logs are pretty-printed for human readability. In production the output is machine-parseable JSON.
Every log entry automatically includes MCP request context (tool name, request ID, workspace ID) when called from within a tool handler:
```json
{
"level": 30,
"time": "2026-04-22T18:30:00.000Z",
"pid": 1234,
"hostname": "workspace-server",
"msg": "rate limit hit",
"tool": "list_workspaces",
"requestId": "req_abc123",
"workspaceId": "ws_xyz789"
}
```
Set `LOG_LEVEL=debug` (level 20) to trace all tool calls and request IDs. Set `LOG_LEVEL=error` (level 50) in CI to suppress informational output.
See [`molecule-mcp-server` PR #6](https://github.com/Molecule-AI/molecule-mcp-server/pull/6) for implementation details.
## Tool Reference
@ -248,3 +274,4 @@ The MCP server handles auth automatically when configured with the correct `MOLE
| "401 Unauthorized" | Token expired or revoked — create a new one |
| Tools not showing | Run `npx @molecule-ai/mcp-server` standalone to check for errors |
| Stale data | MCP server doesn't cache — check platform directly |
| High-volume log output | Set `LOG_LEVEL=error` to suppress info/warn; in production, log output is JSON (machine-parseable) |