graph-tool-call MCP server
Commit 8 Okt 2026Memungkinkan AI client untuk mencari katalog tool yang besar dan hanya mengambil tool target serta producer prasyarat yang dibutuhkan untuk beraksi.
Juga dikenal sebagai SonAIengine/graph-tool-call
- Stars
- 18
- Tools
- 3
- Unduhan, 30 hari
- —
- Versi terbaru
- —
- Commit terakhir
- 8 Okt 2026
- Lisensi
- MIT
Apa itu graph-tool-call MCP?
Memungkinkan AI client untuk mencari katalog tool yang besar dan hanya mengambil tool target serta producer prasyarat yang dibutuhkan untuk beraksi. Server ini menyediakan 3 tool, seperti evidence, token, side.
Ringkasan dibuat dari README proyek
Tentang graph-tool-call MCP server
graph-tool-call memungkinkan Anda mengelola katalog tool yang masif untuk AI agent dengan memperlakukan pengambilan tool sebagai masalah graph, bukan sekadar pencarian kemiripan (similarity search) biasa. Tool ini tidak hanya mengidentifikasi tool utama yang diminta oleh user, tetapi juga tool prasyarat yang diperlukan untuk menghasilkan input yang dibutuhkan bagi target tersebut. Pendekatan ini memastikan agent memiliki rantai eksekusi yang lengkap, seperti menemukan order ID sebelum mencoba memproses refund.
Developer dapat mengintegrasikan logika ini ke dalam aplikasi Python menggunakan ToolGraph atau mendeploy-nya sebagai remote MCP server melalui streamable HTTP. Sistem ini mendukung ingestion dari OpenAPI, GraphQL, MCP tools, dan Python functions. Tool ini bekerja dengan framework utama seperti OpenAI, Anthropic, dan LangChain melalui middleware dan patch khusus. Dengan menyediakan schema yang sesuai dengan budget token, tool ini mengurangi context yang dikirim ke model, yang membantu agar tetap berada dalam batas token sambil tetap menjaga kualitas seleksi yang tinggi.
Apa yang bisa Anda lakukan dengannya
- 1Menyelesaikan rantai dependensi tool yang kompleksAI agent dapat secara otomatis mengidentifikasi bahwa ia perlu memanggil search tool untuk mendapatkan ID sebelum dapat memanggil action tool tertentu.
- 2Mengoptimalkan penggunaan context LLMMengurangi jumlah token yang dikirim ke model dengan menyaring katalog besar menjadi hanya tool yang relevan beserta dependensi yang diperlukan.
- 3Mengolah berbagai spesifikasi APIMengonversi dokumentasi OpenAPI atau GraphQL yang sudah ada menjadi graph terstruktur yang dapat dinavigasi oleh AI.
Ditulis berdasarkan README proyek. Setiap use case diperiksa terhadap kutipan dari README tersebut.
Tool graph-tool-call MCP 3
Terdeteksi dari READMEevidencetokenside
| evidence | — | |
| token | — | |
| side | — |
Pengecekan otomatis
Berjalan 14 Nov 2026Laporan lengkap →Belum diverifikasi. Sebuah server memerlukan skor 80, tidak ada temuan serius, dan data yang cukup untuk kami periksa. Cara kerja checks · Yang perlu diperbaiki untuk menambah poin
- Source repository terhubung dan dapat dijangkausonaiengine/graph-tool-call is public and reachablegithub.com/sonaiengine/graph-tool-call ↗Lolos
- Nama paket mencantumkan repositori ini sebagai sumbernyaTidak diperiksa
- Provenance build (atestasi Sigstore)Tidak diperiksa
- Lisensi open source (OSI)MITspdx.org/licenses/MIT.html ↗Lolos
- Ada commit atau release dalam satu tahun terakhirLast commit or release 2026-10-08github.com/sonaiengine/graph-tool-call ↗Lolos
- OpenSSF ScorecardTidak diperiksa
- Nama package tidak mirip dengan package populerTidak diperiksa
- Tidak ada advisory kritis pada paket atau dependensinyaTidak diperiksa
- Tidak ada instruksi tersembunyi untuk model3 tool descriptions and the README read; no override phrases, credential paths or hidden charactersLolos
- Tools yang bersifat destruktif telah diberi labelNo tool named like delete, drop or execLolos
- Endpoint remote merespons handshake MCPTidak diperiksa
README
Lihat di GitHubgraph-tool-call
Graph-structured retrieval for large LLM tool catalogs.
Find the target tool, the prerequisite tools that produce its inputs, and the smallest schema bundle that fits the planner's token budget.
Documentation · Quickstart · PyPI · Benchmarks
English · 한국어
The Problem
A semantic search for "refund an order" can find refundOrder. That is not
enough when the operation requires an order_id that the user does not have.
A usable candidate set also needs the operation that produces that field:
findOrdersByEmail(email) -> order_id -> refundOrder(order_id)
Large catalogs create a second problem: sending every schema to the model wastes context and can lower selection quality. graph-tool-call treats retrieval as a contract-aware graph problem instead of flat similarity search.
It provides:
- deterministic ingestion from OpenAPI, GraphQL introspection, MCP tools, Python functions, and structured tool catalogs;
- hybrid target retrieval with keyword, graph, optional embedding, and MCP annotation signals;
- evidence-backed target selection and typed prerequisite expansion;
- token-budgeted, contract-projected schemas for the model-facing catalog;
- readiness, failure, and trace metadata for application-side diagnostics;
- adapters for OpenAI, Anthropic, LangChain v1, MCP, Docker, and Kubernetes.
Authentication, tenant policy, approval, and product-specific execution remain in the host application.
See It in 30 Seconds
No model, API key, or network call is required:
uvx graph-tool-call demo dependency-chain
Selected target:
refundOrder(order_id)
Required producer:
findOrdersByEmail(email) -> order_id
evidence: api_contract, openapi_link
Execution order:
1. findOrdersByEmail
2. refundOrder
Planner context:
6 catalog tools -> 2 required tools
estimated tokens: 1476 -> 160 (89% fewer)
This demo runs the real retriever, deterministic target selector, typed dependency closure, and schema admission pipeline.
Installation
The core search and graph package uses only the Python standard library. Optional integrations are installed explicitly:
pip install graph-tool-call
pip install "graph-tool-call[openapi]" # YAML OpenAPI documents
pip install "graph-tool-call[korean]" # Kiwi tokenizer
pip install "graph-tool-call[langchain]" # LangChain v1 middleware
pip install "graph-tool-call[mcp]" # MCP server and proxy
pip install "graph-tool-call[all]" # all optional features
Python 3.10 through 3.14 are tested in CI.
Build and Search
OpenAPI
from graph_tool_call import ToolGraph
graph = ToolGraph.from_url(
"https://petstore3.swagger.io/api/v3/openapi.json",
cache="petstore.graph.json",
)
for result in graph.retrieve_with_scores("create a new pet", top_k=5):
print(result.tool.name, result.score, result.confidence)
OpenAPI ingestion preserves request and response schemas, parameter locations,
content types, security requirements, links, examples, response envelopes, and
typed consumes/produces contracts. Swagger 2.0, OpenAPI 3.0, and OpenAPI 3.1
are supported.
Inspect a collection before exposing it to an agent:
graph-tool-call inspect-openapi ./openapi.json --json
graph-tool-call build-openapi-collection ./openapi.json -o collection.json
The report contains stable readiness issue codes, semantic coverage, and edge quality rather than a single opaque score.
Other sources
from graph_tool_call.ingest import ingest_source
openapi_result = ingest_source(openapi_document)
graphql_result = ingest_source(introspection_result)
mcp_result = ingest_source({"tools": mcp_tools}, format_hint="mcp-tools")
python_result = ingest_source([read_file, write_file])
Every adapter returns normalized ToolSchema objects, capability metadata, and
structured unsupported-feature diagnostics.
Choose an Integration
| Environment | Recommended surface | What graph-tool-call owns |
|---|---|---|
| Python application | ToolGraph / graphify APIs |
ingest, search, evidence, dependency closure |
| OpenAI Responses or Chat Completions | patch_openai |
per-request function-tool filtering |
| Anthropic Messages | patch_anthropic |
per-request tool filtering |
| LangChain v1 | create_tool_selection_middleware |
model-call tool selection |
| Claude Code, Cursor, Windsurf | MCP proxy | many MCP backends behind 3 gateway tools |
| OpenAI Agents, PydanticAI, Google ADK | remote MCP server | protocol-neutral search service |
| Docker or Kubernetes | Streamable HTTP MCP | private deployable service |
See the compatibility matrix for validation boundaries. Protocol compatibility does not imply that every framework or cloud release is tested by this repository.
OpenAI Responses
from graph_tool_call.middleware import patch_openai
patch_openai(client, graph=graph, top_k=5)
response = client.responses.create(
model=model_name,
input="delete a user account",
tools=all_function_tools,
)
Hosted tools such as web search pass through unchanged. The same patch keeps legacy Chat Completions support.
LangChain v1
from langchain.agents import create_agent
from graph_tool_call.langchain import create_tool_selection_middleware
selection = create_tool_selection_middleware(langchain_tools, top_k=5)
agent = create_agent(
model,
tools=langchain_tools,
middleware=[selection],
)
The middleware intersects with tools still allowed by earlier permission or feature-flag middleware; it does not reintroduce filtered tools.
MCP server
graph-tool-call ingest ./openapi.json -o graph.json
graph-tool-call serve \
--graph graph.json \
--transport streamable-http \
--host 127.0.0.1 \
--port 8000
The MCP endpoint is /mcp; HTTP deployments also expose /healthz and
/readyz. Keep remote endpoints private or behind an authenticated gateway.
MCP proxy
graph-tool-call proxy \
--config ./mcp-backends.json \
--transport streamable-http
The proxy accepts local stdio, SSE, and Streamable HTTP backends. In gateway
mode it exposes search_tools, get_tool_schema, and call_backend_tool, then
notifies capable clients when matching backend tools become visible.
Reproducible Evidence
The release headline is deliberately model-free and small enough to replay in CI. On seven curated commerce cases, adding typed producer expansion to the same selected target produced:
| Metric | Target only | Target + graph producers |
|---|---|---|
| Required-producer recall | 14.3% | 100% |
| Candidate plan coverage | 47.6% | 100% |
| Candidate binding support | 14.3% | 100% |
| Target Recall@5 | - | 100% |
The case-level v0.47.0 artifact records fixture hashes, every expected target and producer, and replay commands:
make launch-evidence
make launch-evidence-check
The separate
observability artifact
checks that tracing leaves engine inputs unchanged, replays deterministically,
scrubs secrets, explains every decision, and stays below the documented
5ms/span p95 capture-cost gate:
make observability-evidence-check
This is an engine regression suite, not a population-level estimate of LLM tool-calling accuracy and not a state-of-the-art claim. Larger external comparisons, model-loop experiments, confidence intervals, and known weak cases are reported in Benchmark Results and the paper protocol.
Production Boundary
graph-tool-call is the retrieval and contract layer. A production adapter still owns:
- user and service authentication;
- tenant authorization and approval policy;
- downstream secrets and cookie handling;
- side-effect confirmation, cleanup, and audit retention;
- provider/model lifecycle and final response policy.
Do not store raw credentials in graph artifacts, tool descriptions, trace records, or model-visible arguments.
Documentation
| Start here | Purpose |
|---|---|
| Quickstart | first search, graph, readiness, and execution loop |
| Mental model | understand the pipeline and boundaries |
| OpenAPI ingestion | contract extraction and collection build |
| Target selection | ranking evidence and deterministic guard |
| Integrations | frameworks, MCP, and deployment |
| API reference | stable public Python surface |
| Roadmap | current product and research priorities |
Development
git clone https://github.com/SonAIengine/graph-tool-call.git
cd graph-tool-call
poetry install --with dev --all-extras
poetry run ruff check .
poetry run ruff format --check .
poetry run pytest tests/ -q
See CONTRIBUTING.md and the release checklist.
License
Pertanyaan tentang graph-tool-call MCP
Tool apa yang disediakan graph-tool-call MCP?
Server ini menyediakan 3 tool, termasuk evidence, token, side.
Bagaimana graph-tool-call mengurangi jumlah tool yang dikirim ke model?
Tool ini mengambil tool target beserta producer prasyarat yang bertipe, dan hanya memasukkan schema yang sesuai dengan budget token serta proyeksi kontrak ke dalam katalog yang menghadap ke model.
Format katalog tool apa saja yang bisa digunakan?
Tool ini dapat mengolah OpenAPI, GraphQL introspection, MCP tools, Python functions, dan katalog tool terstruktur.
Apakah ini berjalan sebagai MCP server?
Ya, tool ini dapat menyajikan graph yang telah dibangun melalui streamable HTTP pada endpoint /mcp, dan mode proxy dapat menempatkan banyak backend MCP di balik tiga gateway tools.
Apakah memerlukan dependensi Python tambahan?
Package inti untuk search dan graph hanya menggunakan Python standard library; integrasi seperti OpenAPI, LangChain, dan MCP adalah tambahan opsional.
Apakah ini menangani autentikasi dan otorisasi?
Tidak, ini adalah layer retrieval dan kontrak; aplikasi host yang bertanggung jawab atas autentikasi, kebijakan tenant, dan persetujuan.