Initial commit: hardened DeerFlow factory

Vendored deer-flow upstream (bytedance/deer-flow) plus prompt-injection
hardening:

- New deerflow.security package: content_delimiter, html_cleaner,
  sanitizer (8 layers — invisible chars, control chars, symbols, NFC,
  PUA, tag chars, horizontal whitespace collapse with newline/tab
  preservation, length cap)
- New deerflow.community.searx package: web_search, web_fetch,
  image_search backed by a private SearX instance, every external
  string sanitized and wrapped in <<<EXTERNAL_UNTRUSTED_CONTENT>>>
  delimiters
- All native community web providers (ddg_search, tavily, exa,
  firecrawl, jina_ai, infoquest, image_search) replaced with hard-fail
  stubs that raise NativeWebToolDisabledError at import time, so a
  misconfigured tool.use path fails loud rather than silently falling
  back to unsanitized output
- Native client back-doors (jina_client.py, infoquest_client.py)
  stubbed too
- Native-tool tests quarantined under tests/_disabled_native/
  (collect_ignore_glob via local conftest.py)
- Sanitizer Layer 7 fix: only collapse horizontal whitespace, preserve
  newlines and tabs so list/table structure survives
- Hardened runtime config.yaml references only the searx-backed tools
- Factory overlay (backend/) kept in sync with deer-flow tree as a
  reference / source

See HARDENING.md for the full audit trail and verification steps.
This commit is contained in:
2026-04-12 14:23:57 +02:00
commit 6de0bf9f5b
889 changed files with 173052 additions and 0 deletions

View File

@@ -0,0 +1,68 @@
"""MCP client using langchain-mcp-adapters."""
import logging
from typing import Any
from deerflow.config.extensions_config import ExtensionsConfig, McpServerConfig
logger = logging.getLogger(__name__)
def build_server_params(server_name: str, config: McpServerConfig) -> dict[str, Any]:
"""Build server parameters for MultiServerMCPClient.
Args:
server_name: Name of the MCP server.
config: Configuration for the MCP server.
Returns:
Dictionary of server parameters for langchain-mcp-adapters.
"""
transport_type = config.type or "stdio"
params: dict[str, Any] = {"transport": transport_type}
if transport_type == "stdio":
if not config.command:
raise ValueError(f"MCP server '{server_name}' with stdio transport requires 'command' field")
params["command"] = config.command
params["args"] = config.args
# Add environment variables if present
if config.env:
params["env"] = config.env
elif transport_type in ("sse", "http"):
if not config.url:
raise ValueError(f"MCP server '{server_name}' with {transport_type} transport requires 'url' field")
params["url"] = config.url
# Add headers if present
if config.headers:
params["headers"] = config.headers
else:
raise ValueError(f"MCP server '{server_name}' has unsupported transport type: {transport_type}")
return params
def build_servers_config(extensions_config: ExtensionsConfig) -> dict[str, dict[str, Any]]:
"""Build servers configuration for MultiServerMCPClient.
Args:
extensions_config: Extensions configuration containing all MCP servers.
Returns:
Dictionary mapping server names to their parameters.
"""
enabled_servers = extensions_config.get_enabled_mcp_servers()
if not enabled_servers:
logger.info("No enabled MCP servers found")
return {}
servers_config = {}
for server_name, server_config in enabled_servers.items():
try:
servers_config[server_name] = build_server_params(server_name, server_config)
logger.info(f"Configured MCP server: {server_name}")
except Exception as e:
logger.error(f"Failed to configure MCP server '{server_name}': {e}")
return servers_config