Python example scripts demonstrating how to use InfoMesh as a library.
# Clone and install
git clone https://github.com/dotnetpower/infomesh.git
cd infomesh
uv sync| Script | Description |
|---|---|
| basic_search.py | Search your local index using FTS5 |
| crawl_and_search.py | Crawl a URL, index it, then search |
| hybrid_search.py | Keyword + vector semantic search |
| fetch_page.py | Fetch full text of a web page |
| credit_status.py | Check credit balance and earnings |
| mcp_client.py | Connect to InfoMesh MCP server programmatically |
# Basic search
uv run python examples/basic_search.py "python asyncio"
# Crawl a page and search it
uv run python examples/crawl_and_search.py https://docs.python.org/3/library/asyncio.html
# Hybrid search (requires: uv add chromadb sentence-transformers)
uv run python examples/hybrid_search.py "how to handle errors in async python"
# Fetch a page
uv run python examples/fetch_page.py https://example.com
# Check credits
uv run python examples/credit_status.py
# MCP client (connects to the MCP server as a subprocess)
uv run python examples/mcp_client.pyYou can import InfoMesh components directly into your own Python projects:
from infomesh.config import load_config
from infomesh.index.local_store import LocalStore
from infomesh.search.query import search_local
config = load_config()
store = LocalStore(db_path=config.index.db_path)
result = search_local(store, "your query", limit=10)
for doc in result.documents:
print(f"{doc.title} — {doc.url}")
store.close()