plane_mcp/__main__.py (2)

37-52: Module-level logging configuration runs on import.

Calling configure_json_logging() at module level (line 52) means the logging configuration is applied whenever this module is imported, not just when main() is called. This could affect tests or other code that imports from this module.

Consider moving the call inside main() or guarding it:

Alternative: configure inside main()

-configure_json_logging()
-
-logger = logging.getLogger("fastmcp.plane_mcp")
+logger = logging.getLogger("fastmcp.plane_mcp")


 class ServerMode(Enum):
     ...

 def main() -> None:
     """Run the MCP server."""
+    configure_json_logging()
+
     server_mode = ServerMode.STDIO

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@plane_mcp/__main__.py` around lines 37 - 52, The module currently calls
configure_json_logging() at import time which alters the "fastmcp" logger for
all importers; move the configure_json_logging() invocation into the main()
entry point (or wrap it behind an if __name__ == "__main__": guard) so the
logging mutation only happens when the program is executed, not when the module
is imported; update references to configure_json_logging and ensure main()
invokes it before any FastMCP activity so behavior remains unchanged when run.

---

19-34: Consider including full stack trace in error logging.

The JSONFormatter captures only the exception type and message but omits the traceback. For debugging production issues, the full stack trace is often essential.

Proposed enhancement to include traceback

+import traceback
+
 class JSONFormatter(logging.Formatter):
     """JSON log formatter for structured logging (Datadog, ELK, etc.)."""

     def format(self, record: logging.LogRecord) -> str:
         log_entry = {
             "timestamp": datetime.fromtimestamp(record.created, tz=timezone.utc).isoformat(),
             "level": record.levelname,
             "logger": record.name,
             "message": record.getMessage(),
         }
         if record.exc_info and record.exc_info[1]:
             log_entry["error"] = {
                 "type": type(record.exc_info[1]).__name__,
                 "message": str(record.exc_info[1]),
+                "traceback": traceback.format_exception(*record.exc_info),
             }
         return json.dumps(log_entry)

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@plane_mcp/__main__.py` around lines 19 - 34, The JSONFormatter.format
currently stores only exception type and message from record.exc_info; update
JSONFormatter.format to include the full traceback by formatting record.exc_info
(e.g., call self.formatException(record.exc_info) or use
traceback.format_exception) and add that string as an additional field (e.g.,
log_entry["error"]["traceback"] or "stack") inside the error object; ensure you
only attempt to format when record.exc_info is present and that the traceback
content is JSON-serializable (convert to a single string).

.github/workflows/build_check.yml (1)

14-24: Missing Node.js version setup step.

The workflow relies on the runner's default Node.js version, which can vary and cause inconsistent builds. Add a setup-node action to pin the Node.js version.

Proposed fix

       - name: Checkout code
         uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+          cache: 'npm'

       - name: Install dependencies
         run: npm ci

Also, fetch-depth: 0 fetches the entire git history, which is unnecessary for a build check and slows down checkout. Remove it unless you specifically need commit history.

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In @.github/workflows/build_check.yml around lines 14 - 24, Add a setup-node
step before "Install dependencies" to pin the Node.js version (e.g. use
actions/setup-node@v4 with node-version set to your project's supported version
or package.json engines) so builds are consistent, and remove the unnecessary
fetch-depth: 0 from the "Checkout code" step to avoid fetching full git history
during the build check; update the workflow to place the setup-node step between
the "Checkout code" and "Install dependencies" steps.

plane_mcp/__main__.py (1)

19-34: Consider including full stack trace in error logging.

The JSONFormatter captures exception type and message but omits the full traceback. For debugging production issues, the stack trace is often essential.

♻️ Proposed enhancement to include traceback

+import traceback
+
 class JSONFormatter(logging.Formatter):
     """JSON log formatter for structured logging (Datadog, ELK, etc.)."""

     def format(self, record: logging.LogRecord) -> str:
         log_entry = {
             "timestamp": datetime.fromtimestamp(record.created, tz=timezone.utc).isoformat(),
             "level": record.levelname,
             "logger": record.name,
             "message": record.getMessage(),
         }
         if record.exc_info and record.exc_info[1]:
             log_entry["error"] = {
                 "type": type(record.exc_info[1]).__name__,
                 "message": str(record.exc_info[1]),
+                "traceback": traceback.format_exception(*record.exc_info),
             }
         return json.dumps(log_entry)

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@plane_mcp/__main__.py` around lines 19 - 34, The JSONFormatter currently
records exception type and message but not the full stack trace; update
JSONFormatter.format to capture the traceback when record.exc_info is present
(use the record.exc_info tuple with traceback.format_exception or
traceback.format_exception_only/format_exception to produce a single string) and
add it to the log_entry under error (e.g. error["traceback"]) so structured logs
include the full stack trace; ensure you import and use the traceback module and
fall back safely if exc_info is absent (optionally include record.stack_info if
available).