GitNexus/eval/bridge/mcp_bridge.py
John R. Eakin c68d7975e6
docs: agent development framework, GitHub templates, eval refactor (#479)
* ci: E2E workflow, web typecheck job, pre-commit hook, test suite

CI:
- ci.yml consolidated to reference ci-tests.yml
- ci-quality.yml: add typecheck-web job for gitnexus-web/
- ci-e2e.yml: E2E workflow with dorny/paths-filter (web changes only)
- ci-report.yml: remove dead integration-reports references
- CI gate allows skipped E2E status
- .gitignore: playwright artifacts, eval test artifacts

Pre-commit hook:
- .githooks/pre-commit: typecheck + unit tests for both packages
- Activated via git config core.hooksPath in prepare script

Test infrastructure:
- Vitest + React Testing Library: 58 unit tests
  (graph, server-connection, mermaid, settings, constants, utils, paths)
- Playwright E2E: 5 tests + manual recording harness
- vitest.config from vitest/config, engines.node >= 20
- Playwright artifacts retain-on-failure
- wait-on in devDependencies
- vitest/coverage-v8 aligned with vitest 4.x

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: update gitnexus-web package-lock.json

Reflects devDependency additions (vitest, playwright, wait-on,
@testing-library, etc.) from package.json changes in this PR.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(e2e): add missing process-list-loaded testid, increase CI timeouts

- Add data-testid="process-list-loaded" to ProcessesPanel (E2E tests
  were waiting for an element that didn't exist)
- Increase server connect timeouts from 5s to 10s for slower CI

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(ci): run gitnexus-web unit tests in CI, remove unused variable

- Add gitnexus-web npm ci + vitest run to ci-tests.yml so web unit
  tests are gated by the CI status check (were only running locally)
- Remove unused IS_PLAYWRIGHT_AUTOMATION variable from E2E spec

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(e2e): add process-row testid, wait for networkidle on page load

- Add data-testid="process-row" to ProcessItem component (E2E tests
  referenced it but it didn't exist in the source)
- Use waitUntil: 'networkidle' on page.goto to ensure Vite dev server
  is fully ready before interacting (fixes first-test timeout in CI)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(e2e): add process-view-button and process-highlight-button testids

E2E tests referenced these data-testid attributes but they didn't
exist in ProcessItem. All 6 E2E testids now have matching source
elements: status-ready, process-list-loaded, process-row,
process-view-button, process-highlight-button, server-url-input.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(e2e): remove networkidle — Vite HMR WebSocket prevents it from resolving

networkidle waits for zero network activity for 500ms, but Vite's HMR
WebSocket stays open permanently, causing page.goto to timeout at 60s
on all tests after the first. The explicit toBeVisible waits on UI
elements are sufficient and deterministic.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(e2e): wait for Server button visibility, add CI retry, all 5 tests pass locally

Root cause: test 1 clicked the Server button before React hydrated,
so the tab content never rendered and the input wasn't found.

Fixes:
- Wait for Server button toBeVisible before clicking
- Increase input wait to 15s
- Remove networkidle (Vite HMR WebSocket prevents it from resolving)
- Add retries: 1 in CI for transient cold-start flakiness

Verified locally: all 5 E2E tests pass, 198 unit tests pass, typecheck clean.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(ci): tolerate LadybugDB native crash during analyze step

gitnexus analyze can crash with "double free or corruption" (known
issue #273) during the LadybugDB native addon shutdown. The index is
usually written successfully before the crash. The workflow now:
1. Allows analyze to exit non-zero with a warning
2. Verifies .gitnexus index was actually created
3. Only fails if no index exists (real failure)

All tests verified locally: 198 unit, 5 E2E pass, typecheck clean.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(ci): fix shell quoting in analyze step, simplify to || true

The previous echo string had special characters that broke bash
quoting in GitHub Actions. Simplified to: analyze || true, then
check if .gitnexus exists.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* docs: add agent development framework, GitHub templates, eval refactor

Agent framework (layered docs for AI-assisted contributions):
- AGENTS.md: canonical instructions, impact analysis, MCP tools
- CLAUDE.md: Claude Code-specific deltas and hooks
- GUARDRAILS.md: safety boundaries, non-negotiables, escalation
- ARCHITECTURE.md: monorepo layout, data flow map
- TESTING.md: test structure, commands, categories
- RUNBOOK.md: copy-paste operations for dev/CI/MCP
- llms.txt: minimal LLM context pointer

Editor integration:
- .cursor/index.mdc + rules/100-monorepo.mdc

GitHub templates:
- PR template with areas-touched checkboxes
- Bug report + feature request issue forms

Eval harness:
- Refactored mcp_bridge, tool_registry, constants
- Error sanitization utilities
- Property-based tests via Hypothesis

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix(eval): use format_exception instead of format_exc in sanitize_exception

format_exc() returns the currently handled exception traceback, which
may be unrelated if called outside an active except block. Using
format_exception(type(exc), exc, exc.__traceback__) reliably captures
the passed exception's traceback.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* docs: update CONTRIBUTING.md and TESTING.md for current CI/hook setup

- CONTRIBUTING.md: add gitnexus-web typecheck command, pre-commit hook
  checklist item
- TESTING.md: add gitnexus-web typecheck command, pre-commit hook
  section (husky), update CI integration to list actual workflow files
  (ci-quality, ci-tests, ci-e2e)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* docs: update testing docs to reflect CI/E2E changes from PR #486

- AGENTS.md: update test counts (CLI ~2000 unit, ~1850 integration),
  add gitnexus-web testing section (198 unit, 5 E2E with commands)
- RUNBOOK.md: fix Node requirement to >=20, fix E2E local repro command
- TESTING.md: E2E uses data-testid selectors + real servers, not mocks
- .cursor/rules/100-monorepo.mdc: add web test/E2E commands

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* docs: address context engineering review — deduplicate tokens, expand Cursor rules

- Remove ~100-line gitnexus:start block from CLAUDE.md (was duplicated from AGENTS.md)
- Fix gitnexus:start block inlined inside AGENTS.md Reference Docs bullet (doubled)
- Replace CLAUDE.md scope table with pointer to AGENTS.md (single source of truth)
- Expand .cursor/index.mdc with 5 non-negotiable safety rules for always-on context
- Add .cursor/rules/200-eval.mdc with Python/eval commands (glob-scoped to eval/**)
- Improve llms.txt with priority annotations and descriptions
- Bump version headers to 1.2.0, last-reviewed to 2026-03-24

Saves ~1,400 tokens/session with zero information loss.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Gergő Magyar <gergomagyar@icloud.com>
2026-03-25 06:48:41 +00:00

378 lines
12 KiB
Python

"""
MCP Bridge for GitNexus
Starts the GitNexus MCP server as a subprocess and provides a Python interface
to call MCP tools. Used by the bash wrapper scripts and the augmentation layer..
The bridge communicates with the MCP server via stdio using the JSON-RPC protocol.
"""
import json
import logging
import os
import subprocess
import sys
import threading
import time
from pathlib import Path
from typing import Any
from constants import (
MCP_FIND_GITNEXUS_FALLBACK_TIMEOUT_SECONDS,
MCP_FIND_GITNEXUS_TIMEOUT_SECONDS,
MCP_READ_TIMEOUT_SECONDS,
MCP_STOP_WAIT_SECONDS,
)
from utils.errors import is_debug_enabled, log_safe_exception
logger = logging.getLogger("mcp_bridge")
class MCPBridge:
"""
Manages a GitNexus MCP server subprocess and proxies tool calls to it.
Usage:
bridge = MCPBridge(repo_path="/path/to/repo")
bridge.start()
result = bridge.call_tool("query", {"query": "authentication"})
bridge.stop()
"""
def __init__(self, repo_path: str | None = None):
self.repo_path = repo_path or os.getcwd()
self.process: subprocess.Popen | None = None
self._request_id = 0
self._lock = threading.Lock()
self._started = False
def start(self) -> bool:
"""Start the GitNexus MCP server subprocess."""
if self._started:
return True
try:
# Find gitnexus binary
gitnexus_bin = self._find_gitnexus()
if not gitnexus_bin:
logger.error("GitNexus not found. Install with: npm install -g gitnexus")
return False
self.process = subprocess.Popen(
[gitnexus_bin, "mcp"],
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
cwd=self.repo_path,
text=False,
)
# Send initialize request
init_result = self._send_request("initialize", {
"protocolVersion": "2024-11-05",
"capabilities": {},
"clientInfo": {"name": "gitnexus-eval", "version": "0.1.0"},
})
if init_result is None:
logger.error("MCP server failed to initialize")
self.stop()
return False
# Send initialized notification
self._send_notification("notifications/initialized", {})
self._started = True
logger.info("MCP bridge started successfully")
return True
except Exception as e:
log_safe_exception(logger, "Failed to start MCP bridge", e, include_debug=is_debug_enabled())
self.stop()
return False
def stop(self):
"""Stop the MCP server subprocess."""
if self.process:
try:
if self.process.stdin:
self.process.stdin.close()
if self.process.stdout:
self.process.stdout.close()
if self.process.stderr:
self.process.stderr.close()
self.process.terminate()
self.process.wait(timeout=MCP_STOP_WAIT_SECONDS)
except Exception:
try:
self.process.kill()
except Exception:
pass
self.process = None
self._started = False
def call_tool(self, tool_name: str, arguments: dict[str, Any] | None = None) -> dict[str, Any] | None:
"""
Call a GitNexus MCP tool and return the result.
Returns the tool result content or None on error.
"""
if not self._started:
logger.error("MCP bridge not started")
return None
result = self._send_request("tools/call", {
"name": tool_name,
"arguments": arguments or {},
})
if result is None:
return None
# Extract text content from MCP response
content = result.get("content", [])
if content and isinstance(content, list):
texts = [item.get("text", "") for item in content if item.get("type") == "text"]
return {"text": "\n".join(texts), "raw": content}
return {"text": "", "raw": content}
def list_tools(self) -> list[dict]:
"""List available MCP tools."""
result = self._send_request("tools/list", {})
if result:
return result.get("tools", [])
return []
def read_resource(self, uri: str) -> str | None:
"""Read an MCP resource by URI."""
result = self._send_request("resources/read", {"uri": uri})
if result:
contents = result.get("contents", [])
if contents:
return contents[0].get("text", "")
return None
def _find_gitnexus(self) -> str | None:
"""Find the gitnexus CLI binary."""
# Check if npx is available (preferred - uses local install)
for cmd in ["npx"]:
try:
result = subprocess.run(
[cmd, "gitnexus", "--version"],
capture_output=True,
text=True,
timeout=MCP_FIND_GITNEXUS_TIMEOUT_SECONDS,
cwd=self.repo_path,
)
if result.returncode == 0:
return cmd # Will use "npx gitnexus mcp"
except Exception:
continue
# Check for global install
try:
result = subprocess.run(
["gitnexus", "--version"],
capture_output=True,
text=True,
timeout=MCP_FIND_GITNEXUS_FALLBACK_TIMEOUT_SECONDS,
)
if result.returncode == 0:
return "gitnexus"
except Exception:
pass
return None
def _next_id(self) -> int:
with self._lock:
self._request_id += 1
return self._request_id
def _send_request(self, method: str, params: dict) -> dict | None:
"""Send a JSON-RPC request and wait for response."""
if not self.process or not self.process.stdin or not self.process.stdout:
return None
request_id = self._next_id()
request = {
"jsonrpc": "2.0",
"id": request_id,
"method": method,
"params": params,
}
try:
message = json.dumps(request)
# MCP uses Content-Length header framing
header = f"Content-Length: {len(message.encode('utf-8'))}\r\n\r\n"
self.process.stdin.write(header.encode("utf-8"))
self.process.stdin.write(message.encode("utf-8"))
self.process.stdin.flush()
# Read response
response = self._read_response(timeout=MCP_READ_TIMEOUT_SECONDS)
if response and response.get("id") == request_id:
if "error" in response:
logger.error(f"MCP error: {response['error']}")
return None
return response.get("result")
return None
except Exception as e:
log_safe_exception(logger, "MCP request failed", e, include_debug=is_debug_enabled())
return None
def _send_notification(self, method: str, params: dict):
"""Send a JSON-RPC notification (no response expected)."""
if not self.process or not self.process.stdin:
return
notification = {
"jsonrpc": "2.0",
"method": method,
"params": params,
}
try:
message = json.dumps(notification)
header = f"Content-Length: {len(message.encode('utf-8'))}\r\n\r\n"
self.process.stdin.write(header.encode("utf-8"))
self.process.stdin.write(message.encode("utf-8"))
self.process.stdin.flush()
except Exception as e:
log_safe_exception(logger, "MCP notification failed", e, include_debug=is_debug_enabled())
def _read_content_length(self, deadline: float) -> int | None:
"""Read Content-Length header, returning the byte length or None."""
if not self.process or not self.process.stdout:
return None
header_line = b""
while time.time() < deadline:
byte = self.process.stdout.read(1)
if not byte:
return None
header_line += byte
if header_line.endswith(b"\r\n\r\n") or header_line.endswith(b"\n\n"):
break
if not header_line:
return None
header_str = header_line.decode("utf-8").strip()
for line in header_str.split("\r\n"):
if line.lower().startswith("content-length:"):
try:
return int(line.split(":", 1)[1].strip())
except (ValueError, IndexError):
return None
return None
def _read_body(self, content_length: int, deadline: float) -> bytes | None:
"""Read a response body of the expected length before deadline."""
if not self.process or not self.process.stdout:
return None
remaining = content_length
chunks: list[bytes] = []
while remaining > 0 and time.time() < deadline:
chunk = self.process.stdout.read(remaining)
if not chunk:
return None
chunks.append(chunk)
remaining -= len(chunk)
if remaining > 0:
return None
return b"".join(chunks)
def _read_response(self, timeout: float = MCP_READ_TIMEOUT_SECONDS) -> dict | None:
"""Read a JSON-RPC response from the MCP server."""
if not self.process or not self.process.stdout:
return None
try:
deadline = time.time() + timeout
while time.time() < deadline:
content_length = self._read_content_length(deadline)
if content_length is None:
continue
body = self._read_body(content_length, deadline)
if not body:
return None
message = json.loads(body.decode("utf-8"))
# Skip notifications (no id), return responses
if "id" in message:
return message
return None
except Exception as e:
log_safe_exception(logger, "Error reading MCP response", e, include_debug=is_debug_enabled())
return None
class MCPToolCLI:
"""
CLI wrapper that exposes MCP tools as simple command-line calls.
Used by the bash wrapper scripts inside Docker containers.
Usage from bash:
python -m bridge.mcp_bridge query '{"query": "authentication"}'
python -m bridge.mcp_bridge context '{"name": "validateUser"}'
"""
def __init__(self):
self.bridge = MCPBridge()
def run(self, tool_name: str, args_json: str = "{}") -> int:
"""Run a single tool call and print the result."""
try:
args = json.loads(args_json)
except json.JSONDecodeError:
# Try to parse as simple key=value pairs
args = self._parse_simple_args(args_json)
if not self.bridge.start():
print("ERROR: Failed to start GitNexus MCP bridge", file=sys.stderr)
return 1
try:
result = self.bridge.call_tool(tool_name, args)
if result:
print(result.get("text", ""))
return 0
else:
print("No results", file=sys.stderr)
return 1
finally:
self.bridge.stop()
@staticmethod
def _parse_simple_args(args_str: str) -> dict:
"""Parse 'key=value key2=value2' style arguments."""
args = {}
for part in args_str.split():
if "=" in part:
key, value = part.split("=", 1)
args[key] = value
return args
if __name__ == "__main__":
if len(sys.argv) < 2:
print("Usage: python -m bridge.mcp_bridge <tool_name> [args_json]", file=sys.stderr)
print("Tools: query, context, impact, cypher, list_repos, detect_changes, rename", file=sys.stderr)
sys.exit(1)
tool = sys.argv[1]
args_json = sys.argv[2] if len(sys.argv) > 2 else "{}"
cli = MCPToolCLI()
sys.exit(cli.run(tool, args_json))