This commit is contained in:
2026-04-12 23:56:49 -05:00
parent 50b4c1c905
commit 1511423057

View File

@@ -1,10 +1,29 @@
import os
os.environ["MCP_ALLOW_ALL_ORIGINS"] = "1"
import json
import re
from pathlib import Path
from sentence_transformers import SentenceTransformer
from qdrant_client import QdrantClient
from qdrant_client.models import Filter, FieldCondition, MatchValue
from mcp.server.fastmcp import FastMCP
# Add this right after the FastMCP import, before anything else
from mcp.server import streamable_http
streamable_http.ALLOWED_ORIGINS = None # try this first
# If that doesn't work, patch the actual check function:
import mcp.server.streamable_http as _sh
_sh.is_valid_origin = lambda origin, allowed: True
import uvicorn
from starlette.middleware.cors import CORSMiddleware
from starlette.middleware.base import BaseHTTPMiddleware
from starlette.requests import Request
from mcp.server.transport_security import TransportSecuritySettings, TransportSecurityMiddleware
# Monkey-patch to disable DNS rebinding protection entirely
TransportSecurityMiddleware.__init__ = lambda self, settings=None: setattr(
self, "settings", TransportSecuritySettings(enable_dns_rebinding_protection=False)
)
# ── Paths ──────────────────────────────────────────────────────────────────
project_root = Path(__file__).resolve().parent
@@ -12,13 +31,13 @@ project_root = Path(__file__).resolve().parent
# ── Models / Clients ───────────────────────────────────────────────────────
model = SentenceTransformer("nomic-ai/nomic-embed-text-v1.5", trust_remote_code=True)
qdrant = QdrantClient(path=str(project_root / "data" / "qdrant_local"))
qdrant = QdrantClient(path=str(project_root / "data" / "qdrant_local"))
COLLECTION = "apush_chunks"
with open(project_root / "data" / "processed" / "parent_lookup.json") as f:
parent_lookup = json.load(f)
# ── Config (same as notebook) ──────────────────────────────────────────────
# ── Config ─────────────────────────────────────────────────────────────────
TOP_K = 10
SYSTEM_PROMPT = """You are an expert AP US History tutor helping a student ace their APUSH exam.
@@ -50,7 +69,7 @@ def embed_query(query: str) -> list[float]:
normalize_embeddings=True,
).tolist()
# ── Retrieve (same as notebook) ────────────────────────────────────────────
# ── Retrieve ───────────────────────────────────────────────────────────────
def retrieve(query: str) -> dict:
hits = qdrant.query_points(
collection_name=COLLECTION,
@@ -71,7 +90,6 @@ def retrieve(query: str) -> dict:
else:
confidence = "LOW"
# Deduplicate by parent_id
seen_parents = set()
unique_hits = []
for h in hits:
@@ -84,10 +102,9 @@ def retrieve(query: str) -> dict:
sources = []
for h in unique_hits:
pid = h.payload["parent_id"]
parts = parent_lookup.get(pid, [])
pid = h.payload["parent_id"]
parts = parent_lookup.get(pid, [])
full_text = "\n\n".join(p["text"] for p in parts)
sources.append({
"score": h.score,
"chapter_num": h.payload["chapter_num"],
@@ -104,6 +121,14 @@ def retrieve(query: str) -> dict:
"sources": sources,
}
# ── Origin bypass middleware ────────────────────────────────────────────────
class AllowAllOriginsMiddleware(BaseHTTPMiddleware):
async def dispatch(self, request: Request, call_next):
# Spoof origin so FastMCP's internal check passes
request._headers = request.headers.mutablecopy()
request._headers["origin"] = "http://127.0.0.1:11434"
return await call_next(request)
# ── MCP Server ─────────────────────────────────────────────────────────────
mcp = FastMCP("APUSH Tutor")
@@ -141,10 +166,8 @@ def system_prompt() -> str:
# ── Run ────────────────────────────────────────────────────────────────────
if __name__ == "__main__":
import uvicorn
from starlette.middleware.cors import CORSMiddleware
app = mcp.streamable_http_app()
app.add_middleware(AllowAllOriginsMiddleware)
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
@@ -152,4 +175,5 @@ if __name__ == "__main__":
allow_headers=["*"],
)
print("Starting APUSH MCP server on http://127.0.0.1:52437/mcp")
uvicorn.run(app, host="127.0.0.1", port=52437)