Spaces:
Sleeping
Sleeping
File size: 20,388 Bytes
15de73a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 |
#!/usr/bin/env python3
"""
Gradio MCP Server for Congressional Bioguide profiles.
Provides search and analysis capabilities via Gradio interface.
"""
import gradio as gr
import sqlite3
import json
import os
import warnings
from typing import List, Dict, Any
import numpy as np
from sentence_transformers import SentenceTransformer
import faiss
import pickle
from pathlib import Path
# Suppress warnings
warnings.filterwarnings('ignore')
os.environ['TOKENIZERS_PARALLELISM'] = 'false'
# Initialize global resources
SCRIPT_DIR = Path(__file__).parent.absolute()
DB_PATH = str(SCRIPT_DIR / "congress.db")
FAISS_INDEX_PATH = str(SCRIPT_DIR / "congress_faiss.index")
BIO_IDS_PATH = str(SCRIPT_DIR / "congress_bio_ids.pkl")
# Global state
model = None
faiss_index = None
bio_id_mapping = None
def initialize_search_index():
"""Initialize the semantic search components."""
global model, faiss_index, bio_id_mapping
try:
if Path(FAISS_INDEX_PATH).exists() and Path(BIO_IDS_PATH).exists():
print(f"Loading FAISS index from: {FAISS_INDEX_PATH}")
model = SentenceTransformer('all-MiniLM-L6-v2')
faiss_index = faiss.read_index(FAISS_INDEX_PATH)
with open(BIO_IDS_PATH, "rb") as f:
bio_id_mapping = pickle.load(f)
print(f"✓ Loaded {faiss_index.ntotal} embeddings")
return True
else:
print(f"FAISS index not found. Semantic search will be unavailable.")
return False
except Exception as e:
print(f"Error loading search index: {e}")
return False
def get_db_connection():
"""Get a database connection."""
return sqlite3.connect(DB_PATH)
def execute_query(query: str, params: tuple = ()) -> List[Dict[str, Any]]:
"""Execute a SQL query and return results as list of dicts."""
conn = get_db_connection()
conn.row_factory = sqlite3.Row
cursor = conn.cursor()
cursor.execute(query, params)
results = [dict(row) for row in cursor.fetchall()]
conn.close()
return results
# Initialize search index on startup
print("Initializing Congressional Bioguide MCP Server...")
initialize_search_index()
# MCP Tool Functions with decorators
@gr.mcp.tool()
def search_by_name(family_name: str = "", given_name: str = "", limit: int = 10) -> str:
"""
Search for Congressional members by name.
Args:
family_name: Last name to search for (partial match)
given_name: First name to search for (partial match)
limit: Maximum number of results to return (default: 10)
Returns:
JSON string with search results including bio_id, name, birth/death dates, party, state
"""
try:
conditions = []
params = []
if family_name:
conditions.append("LOWER(m.unaccented_family_name) LIKE LOWER(?)")
params.append(f"%{family_name}%")
if given_name:
conditions.append("LOWER(m.unaccented_given_name) LIKE LOWER(?)")
params.append(f"%{given_name}%")
if not conditions:
return json.dumps({"error": "Please provide at least family_name or given_name"})
query = f"""
SELECT DISTINCT m.bio_id, m.given_name, m.middle_name, m.family_name,
m.birth_date, m.death_date,
j.party, j.region_code, j.job_name, j.congress_number
FROM members m
LEFT JOIN job_positions j ON m.bio_id = j.bio_id
WHERE {' AND '.join(conditions)}
ORDER BY m.family_name, m.given_name
LIMIT ?
"""
params.append(limit)
results = execute_query(query, tuple(params))
return json.dumps({"count": len(results), "results": results}, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def search_by_party(party: str, congress_number: int = None) -> str:
"""
Search for Congressional members by political party.
Args:
party: Party name (e.g., 'Republican', 'Democrat', 'Whig')
congress_number: Optional Congress number to filter by (e.g., 117)
Returns:
JSON string with members from the specified party
"""
try:
if congress_number:
query = """
SELECT DISTINCT m.bio_id, m.given_name, m.family_name, m.birth_date, m.death_date,
j.party, j.region_code, j.job_name, j.congress_number
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
WHERE j.party = ? AND j.congress_number = ?
ORDER BY m.family_name, m.given_name
LIMIT 100
"""
results = execute_query(query, (party, congress_number))
else:
query = """
SELECT DISTINCT m.bio_id, m.given_name, m.family_name, m.birth_date, m.death_date,
j.party, j.region_code, j.job_name, j.congress_number
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
WHERE j.party = ?
ORDER BY m.family_name, m.given_name
LIMIT 100
"""
results = execute_query(query, (party,))
return json.dumps({"count": len(results), "party": party, "results": results}, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def search_by_state(state_code: str, congress_number: int = None) -> str:
"""
Search for Congressional members by state.
Args:
state_code: Two-letter state code (e.g., 'CA', 'NY', 'TX')
congress_number: Optional Congress number to filter by
Returns:
JSON string with members from the specified state
"""
try:
state_code = state_code.upper()
if congress_number:
query = """
SELECT DISTINCT m.bio_id, m.given_name, m.family_name, m.birth_date, m.death_date,
j.party, j.region_code, j.job_name, j.congress_number
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
WHERE j.region_code = ? AND j.congress_number = ?
ORDER BY m.family_name, m.given_name
LIMIT 100
"""
results = execute_query(query, (state_code, congress_number))
else:
query = """
SELECT DISTINCT m.bio_id, m.given_name, m.family_name, m.birth_date, m.death_date,
j.party, j.region_code, j.job_name, j.congress_number
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
WHERE j.region_code = ?
ORDER BY m.family_name, m.given_name
LIMIT 100
"""
results = execute_query(query, (state_code,))
return json.dumps({"count": len(results), "state": state_code, "results": results}, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def semantic_search_biography(query: str, top_k: int = 5) -> str:
"""
Perform AI-powered semantic search on member biographies using natural language.
Args:
query: Natural language query (e.g., 'lawyers who became judges', 'Civil War veterans')
top_k: Number of results to return (default: 5, max: 20)
Returns:
JSON string with matching members and their similarity scores
"""
try:
if not all([model, faiss_index, bio_id_mapping]):
return json.dumps({"error": "Semantic search is not available. FAISS index not loaded."})
# Limit top_k
top_k = min(max(1, top_k), 20)
# Encode query
query_embedding = model.encode([query])[0].astype('float32')
query_embedding = query_embedding.reshape(1, -1)
faiss.normalize_L2(query_embedding)
# Search
scores, indices = faiss_index.search(query_embedding, top_k)
# Get profiles
results = []
for idx, score in zip(indices[0], scores[0]):
if idx < len(bio_id_mapping):
bio_id = bio_id_mapping[idx]
member_query = """
SELECT m.bio_id, m.given_name, m.middle_name, m.family_name,
m.birth_date, m.death_date, m.profile_text,
j.party, j.region_code, j.job_name, j.congress_number
FROM members m
LEFT JOIN job_positions j ON m.bio_id = j.bio_id
WHERE m.bio_id = ?
LIMIT 1
"""
member_data = execute_query(member_query, (bio_id,))
if member_data:
member = member_data[0]
# Truncate profile_text for response
if member.get('profile_text'):
member['profile_text'] = member['profile_text'][:500] + "..."
member['similarity_score'] = float(score)
results.append(member)
return json.dumps({"query": query, "count": len(results), "results": results}, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def get_member_profile(bio_id: str) -> str:
"""
Get complete profile for a specific member by their Bioguide ID.
Args:
bio_id: Bioguide ID (e.g., 'L000313' for John Lewis, 'W000374')
Returns:
JSON string with complete member profile including positions and relationships
"""
try:
bio_id = bio_id.upper()
conn = get_db_connection()
conn.row_factory = sqlite3.Row
cursor = conn.cursor()
cursor.execute("SELECT * FROM members WHERE bio_id = ?", (bio_id,))
member = cursor.fetchone()
if not member:
conn.close()
return json.dumps({"error": f"No member found with bio_id: {bio_id}"})
profile = dict(member)
# Get job positions
cursor.execute("SELECT * FROM job_positions WHERE bio_id = ? ORDER BY start_date", (bio_id,))
profile['job_positions'] = [dict(row) for row in cursor.fetchall()]
# Get relationships
cursor.execute("SELECT * FROM relationships WHERE bio_id = ?", (bio_id,))
profile['relationships'] = [dict(row) for row in cursor.fetchall()]
# Get creative works
cursor.execute("SELECT * FROM creative_works WHERE bio_id = ?", (bio_id,))
profile['creative_works'] = [dict(row) for row in cursor.fetchall()]
conn.close()
return json.dumps(profile, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def count_members_by_party(filter_congress: int = None) -> str:
"""
Count members by political party.
Args:
filter_congress: Optional Congress number to filter by (e.g., 117)
Returns:
JSON string with member counts grouped by party
"""
try:
if filter_congress:
query = """
SELECT j.party as party, COUNT(DISTINCT m.bio_id) as count
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
WHERE j.congress_number = ?
GROUP BY j.party
ORDER BY count DESC
"""
results = execute_query(query, (filter_congress,))
else:
query = """
SELECT j.party as party, COUNT(DISTINCT m.bio_id) as count
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
GROUP BY j.party
ORDER BY count DESC
"""
results = execute_query(query)
total = sum(r['count'] for r in results)
return json.dumps({"total_members": total, "by_party": results}, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def count_members_by_state(filter_congress: int = None) -> str:
"""
Count members by state.
Args:
filter_congress: Optional Congress number to filter by
Returns:
JSON string with member counts grouped by state
"""
try:
if filter_congress:
query = """
SELECT j.region_code as state, COUNT(DISTINCT m.bio_id) as count
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
WHERE j.congress_number = ?
GROUP BY j.region_code
ORDER BY count DESC
"""
results = execute_query(query, (filter_congress,))
else:
query = """
SELECT j.region_code as state, COUNT(DISTINCT m.bio_id) as count
FROM members m
JOIN job_positions j ON m.bio_id = j.bio_id
GROUP BY j.region_code
ORDER BY count DESC
"""
results = execute_query(query)
total = sum(r['count'] for r in results)
return json.dumps({"total_members": total, "by_state": results}, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def execute_sql_query(query: str) -> str:
"""
Execute a custom SQL SELECT query against the Congressional database (READ-ONLY).
Args:
query: SQL SELECT query to execute
Returns:
JSON string with query results
"""
try:
# Security: only allow SELECT queries
if not query.strip().upper().startswith("SELECT"):
return json.dumps({"error": "Only SELECT queries are allowed"})
results = execute_query(query)
return json.dumps({"count": len(results), "results": results}, indent=2)
except Exception as e:
return json.dumps({"error": str(e)})
@gr.mcp.tool()
def get_database_schema() -> str:
"""
Get the database schema showing all tables and columns available for querying.
Returns:
JSON string with database schema information
"""
schema_info = {
"tables": {
"members": {
"description": "Main table with member biographical information",
"columns": [
"bio_id (PRIMARY KEY) - Bioguide ID",
"family_name - Last name",
"given_name - First name",
"middle_name - Middle name",
"birth_date - Birth date (YYYY-MM-DD)",
"death_date - Death date (YYYY-MM-DD)",
"profile_text - Full biography text"
]
},
"job_positions": {
"description": "Congressional positions held by members",
"columns": [
"bio_id (FOREIGN KEY) - References members",
"job_name - Position title (Representative, Senator)",
"start_date - Start date of position",
"end_date - End date of position",
"congress_number - Congress number (e.g., 117)",
"party - Party affiliation",
"region_code - State/region code (e.g., 'CA', 'NY')"
]
},
"relationships": {
"description": "Family relationships between members",
"columns": ["bio_id", "related_bio_id", "relationship_type"]
},
"creative_works": {
"description": "Publications and creative works by members",
"columns": ["bio_id", "citation_text"]
}
}
}
return json.dumps(schema_info, indent=2)
# Create Gradio Interfaces for each tool
demo = gr.TabbedInterface(
[
# Search by Name
gr.Interface(
fn=search_by_name,
inputs=[
gr.Textbox(label="Family Name (Last Name)", placeholder="e.g., Lincoln"),
gr.Textbox(label="Given Name (First Name)", placeholder="e.g., Abraham"),
gr.Slider(minimum=1, maximum=50, value=10, step=1, label="Max Results")
],
outputs=gr.JSON(label="Search Results"),
title="Search by Name",
description="Search for Congressional members by their first or last name."
),
# Search by Party
gr.Interface(
fn=search_by_party,
inputs=[
gr.Textbox(label="Party Name", placeholder="e.g., Republican, Democrat, Whig"),
gr.Number(label="Congress Number (optional)", value=None, precision=0)
],
outputs=gr.JSON(label="Search Results"),
title="Search by Party",
description="Find members by political party affiliation."
),
# Search by State
gr.Interface(
fn=search_by_state,
inputs=[
gr.Textbox(label="State Code", placeholder="e.g., CA, NY, TX"),
gr.Number(label="Congress Number (optional)", value=None, precision=0)
],
outputs=gr.JSON(label="Search Results"),
title="Search by State",
description="Find members by the state they represented."
),
# Semantic Search
gr.Interface(
fn=semantic_search_biography,
inputs=[
gr.Textbox(label="Search Query", placeholder="e.g., 'lawyers who became judges' or 'Civil War veterans'", lines=3),
gr.Slider(minimum=1, maximum=20, value=5, step=1, label="Number of Results")
],
outputs=gr.JSON(label="Search Results"),
title="AI Semantic Search",
description="Use natural language to search biographies. Find members by career, background, or accomplishments."
),
# Get Member Profile
gr.Interface(
fn=get_member_profile,
inputs=gr.Textbox(label="Bioguide ID", placeholder="e.g., L000313 (John Lewis)"),
outputs=gr.JSON(label="Member Profile"),
title="Get Member Profile",
description="Get complete profile for a specific member using their Bioguide ID."
),
# Count by Party
gr.Interface(
fn=count_members_by_party,
inputs=gr.Number(label="Filter by Congress Number (optional)", value=None, precision=0),
outputs=gr.JSON(label="Party Counts"),
title="Count by Party",
description="Get member counts grouped by political party."
),
# Count by State
gr.Interface(
fn=count_members_by_state,
inputs=gr.Number(label="Filter by Congress Number (optional)", value=None, precision=0),
outputs=gr.JSON(label="State Counts"),
title="Count by State",
description="Get member counts grouped by state."
),
# SQL Query
gr.Interface(
fn=execute_sql_query,
inputs=gr.Textbox(label="SQL Query", placeholder="SELECT * FROM members LIMIT 10", lines=3),
outputs=gr.JSON(label="Query Results"),
title="Execute SQL",
description="Execute custom SQL SELECT queries (read-only)."
),
# Database Schema
gr.Interface(
fn=get_database_schema,
inputs=None,
outputs=gr.JSON(label="Database Schema"),
title="Database Schema",
description="View the database structure and available tables/columns."
),
],
tab_names=[
"Search by Name",
"Search by Party",
"Search by State",
"AI Semantic Search",
"Member Profile",
"Count by Party",
"Count by State",
"Execute SQL",
"Database Schema"
],
title="🏛️ Congressional Bioguide MCP Server",
theme=gr.themes.Soft()
)
if __name__ == "__main__":
demo.launch(mcp_server=True)
|