graphwalk 0.1.0__tar.gz

graphwalk-0.1.0/LICENSE

@@ -0,0 +1,177 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS

graphwalk-0.1.0/PKG-INFO

@@ -0,0 +1,151 @@
+Metadata-Version: 2.4
+Name: graphwalk
+Version: 0.1.0
+Summary: Cheap, confidence-scored question answering over knowledge graphs: each hop is a calibrated classification decision. Library, CLI, and MCP server.
+Keywords: knowledge graph,kgqa,graph traversal,mcp,llm agents,calibration,retrieval
+Author: Manvir Chakal
+License-Expression: Apache-2.0
+License-File: LICENSE
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Typing :: Typed
+Requires-Dist: networkx>=3.4
+Requires-Dist: numpy>=2.1
+Requires-Dist: pydantic>=2.12
+Requires-Dist: pydantic-settings>=2.6
+Requires-Dist: typer>=0.15
+Requires-Dist: typesafe-sdk>=0.7.2,<0.8
+Requires-Dist: fastembed>=0.7 ; extra == 'embeddings'
+Requires-Dist: httpx>=0.28 ; extra == 'eval'
+Requires-Dist: pyarrow>=17 ; extra == 'eval'
+Requires-Dist: litellm>=1.60 ; extra == 'llm'
+Requires-Dist: mcp>=1.20 ; extra == 'mcp'
+Requires-Dist: pyjwt[crypto]>=2.10 ; extra == 'mcp'
+Requires-Dist: uvicorn>=0.34 ; extra == 'mcp'
+Requires-Python: >=3.12
+Project-URL: Changelog, https://github.com/manvirchakal/graphwalk/blob/main/CHANGELOG.md
+Project-URL: Documentation, https://manvirchakal.github.io/graphwalk/
+Project-URL: Homepage, https://github.com/manvirchakal/graphwalk
+Project-URL: Issues, https://github.com/manvirchakal/graphwalk/issues
+Project-URL: Source, https://github.com/manvirchakal/graphwalk
+Provides-Extra: embeddings
+Provides-Extra: eval
+Provides-Extra: llm
+Provides-Extra: mcp
+Description-Content-Type: text/markdown
+
+# graphwalk
+
+[![CI](https://github.com/manvirchakal/graphwalk/actions/workflows/ci.yml/badge.svg)](https://github.com/manvirchakal/graphwalk/actions/workflows/ci.yml)
+[![PyPI](https://img.shields.io/pypi/v/graphwalk)](https://pypi.org/project/graphwalk/)
+[![Docs](https://img.shields.io/badge/docs-github.io-blue)](https://manvirchakal.github.io/graphwalk/)
+[![License](https://img.shields.io/badge/license-Apache--2.0-green)](https://github.com/manvirchakal/graphwalk/blob/main/LICENSE)
+
+**Cheap, confidence-scored question answering over a knowledge graph you already
+have.** Python library, CLI, and MCP server for agents.
+
+graphwalk answers a question by walking the graph from the entities it names. Each hop
+is a **constrained classification decision**: the current node's relations, plus
+`STOP`, are the options, and a fast decision model returns a probability for each. So
+every answer comes with a path and a **confidence**, and you can escalate the unsure
+ones to a stronger model.
+
+```bash
+pip install graphwalk
+graphwalk import kg.nt --graph kg.db          # triples: CSV, TSV, JSONL, N-Triples; no key needed
+graphwalk query kg.db "Where was the director of Inception born?" --escalate-below 0.9
+```
+
+## When to use it (measured)
+
+Every number below comes from an experiment in [`docs/`](https://github.com/manvirchakal/graphwalk/tree/main/docs/), on samples of 30–300
+questions. Treat them as directions, not guarantees.
+
+| Situation | What we found |
+|---|---|
+| **An agent exploring a KG** with graph tools (MCP) | Adding graphwalk's `walk` tool cut a strong agent's cost by **25%** (95% CI 7–45%) at equal accuracy (0.75 vs 0.71 F1, n=30, WebQSP). Text search over the same facts was as accurate and a little cheaper, but slower. |
+| **Existing KG, cost or latency first**, or you need a per-answer confidence | `walk` is ~6× cheaper and ~4× faster than an LLM writing the query, and its confidence ranks answers well (AUROC 0.92 on WebQSP, up to 0.97 on MetaQA). |
+| **Existing KG, best accuracy** | Have an LLM write the query instead: 0.66 vs 0.49 F1 on a 5,419-relation Freebase graph. |
+| Questions with constraints or superlatives (CWQ) | Not graphwalk. Every system we tried scored about 0.3. |
+| **Questions over documents** | Not graphwalk. Multi-step RAG wins by 10–19 F1. Text ingestion exists but is experimental. |
+
+graphwalk does not claim to be more accurate than an LLM. What it offers is a cheap first
+answer that **knows when it might be wrong**.
+
+## Quickstart
+
+Decisions use the Jev model, through [OpenRouter](https://openrouter.ai)
+(`OPENROUTER_API_KEY`) or TypeSafe (`TYPESAFE_API_KEY`). Without either,
+`GRAPHWALK_DECISION_FALLBACK=llm` lets any chat model decide (slower, not calibrated).
+
+```python
+import asyncio
+from graphwalk import Index
+
+
+async def main() -> None:
+    async with Index.open("kg.db", escalate_below=0.9) as index:
+        await index.import_triples("kg.nt")  # idempotent
+        result = await index.walk("Where was the director of Inception born?")
+        print(result.best.names, result.confidence, result.escalated, result.cost_usd)
+
+
+asyncio.run(main())
+```
+
+A runnable example over a 15-triple movie graph is in
+[`examples/kg/`](https://github.com/manvirchakal/graphwalk/blob/main/examples/kg/walk.py).
+
+## For agents (MCP)
+
+```bash
+pip install 'graphwalk[mcp]'
+claude mcp add --env OPENROUTER_API_KEY=sk-or-... graphwalk -- graphwalk mcp --db kg.db
+```
+
+Tools: `walk` (answers, paths, confidence), `neighbors`, `get_node`, `status`, plus
+`locate`/`read`/`ingest` for text. The usual loop: call `walk` first; at confidence 0.9
+or above, take the answer; below it, check it with `neighbors` or escalate. The server
+also runs over HTTP with bearer or OAuth 2.1 auth, and ships as a container. See
+[MCP setup](https://manvirchakal.github.io/graphwalk/agents/).
+
+The package carries its own guide for coding agents: `graphwalk guide` (or
+`graphwalk guide --skill` for a `SKILL.md`), and [`llms.txt`](https://github.com/manvirchakal/graphwalk/blob/main/llms.txt).
+
+## Install options
+
+| Extra | Enables |
+|---|---|
+| (none) | import, walk, query, the CLI |
+| `mcp` | the MCP server (stdio and HTTP) |
+| `llm` | LiteLLM: escalation, the LLM decider, text ingestion |
+| `embeddings` | local fastembed embedder (prefilter, dense `locate`) |
+| `eval` | dataset loaders for the benchmarks |
+
+## Documentation
+
+- [Docs site](https://manvirchakal.github.io/graphwalk/): quickstart, agents and MCP,
+  configuration, text ingestion, evidence.
+- [Results](https://github.com/manvirchakal/graphwalk/blob/main/docs/evidence.md): every experiment, with CIs, costs, and the runs behind it.
+- [Design record](https://github.com/manvirchakal/graphwalk/blob/main/docs/design.md) and [roadmap](https://github.com/manvirchakal/graphwalk/blob/main/roadmap.md).
+
+## Status
+
+Alpha (v0.1). The public API is what `graphwalk` exports; submodules may change between
+minor versions. Results are from small samples, mostly one seed. The decision model
+(Jev) is a hosted model from TypeSafe; this project has no affiliation with TypeSafe,
+and the LLM-decider fallback keeps it usable without Jev.
+
+## Development
+
+See [CONTRIBUTING.md](https://github.com/manvirchakal/graphwalk/blob/main/CONTRIBUTING.md).
+
+## License
+
+Apache-2.0. See [LICENSE](https://github.com/manvirchakal/graphwalk/blob/main/LICENSE).

graphwalk-0.1.0/README.md

@@ -0,0 +1,109 @@
+# graphwalk
+
+[![CI](https://github.com/manvirchakal/graphwalk/actions/workflows/ci.yml/badge.svg)](https://github.com/manvirchakal/graphwalk/actions/workflows/ci.yml)
+[![PyPI](https://img.shields.io/pypi/v/graphwalk)](https://pypi.org/project/graphwalk/)
+[![Docs](https://img.shields.io/badge/docs-github.io-blue)](https://manvirchakal.github.io/graphwalk/)
+[![License](https://img.shields.io/badge/license-Apache--2.0-green)](https://github.com/manvirchakal/graphwalk/blob/main/LICENSE)
+
+**Cheap, confidence-scored question answering over a knowledge graph you already
+have.** Python library, CLI, and MCP server for agents.
+
+graphwalk answers a question by walking the graph from the entities it names. Each hop
+is a **constrained classification decision**: the current node's relations, plus
+`STOP`, are the options, and a fast decision model returns a probability for each. So
+every answer comes with a path and a **confidence**, and you can escalate the unsure
+ones to a stronger model.
+
+```bash
+pip install graphwalk
+graphwalk import kg.nt --graph kg.db          # triples: CSV, TSV, JSONL, N-Triples; no key needed
+graphwalk query kg.db "Where was the director of Inception born?" --escalate-below 0.9
+```
+
+## When to use it (measured)
+
+Every number below comes from an experiment in [`docs/`](https://github.com/manvirchakal/graphwalk/tree/main/docs/), on samples of 30–300
+questions. Treat them as directions, not guarantees.
+
+| Situation | What we found |
+|---|---|
+| **An agent exploring a KG** with graph tools (MCP) | Adding graphwalk's `walk` tool cut a strong agent's cost by **25%** (95% CI 7–45%) at equal accuracy (0.75 vs 0.71 F1, n=30, WebQSP). Text search over the same facts was as accurate and a little cheaper, but slower. |
+| **Existing KG, cost or latency first**, or you need a per-answer confidence | `walk` is ~6× cheaper and ~4× faster than an LLM writing the query, and its confidence ranks answers well (AUROC 0.92 on WebQSP, up to 0.97 on MetaQA). |
+| **Existing KG, best accuracy** | Have an LLM write the query instead: 0.66 vs 0.49 F1 on a 5,419-relation Freebase graph. |
+| Questions with constraints or superlatives (CWQ) | Not graphwalk. Every system we tried scored about 0.3. |
+| **Questions over documents** | Not graphwalk. Multi-step RAG wins by 10–19 F1. Text ingestion exists but is experimental. |
+
+graphwalk does not claim to be more accurate than an LLM. What it offers is a cheap first
+answer that **knows when it might be wrong**.
+
+## Quickstart
+
+Decisions use the Jev model, through [OpenRouter](https://openrouter.ai)
+(`OPENROUTER_API_KEY`) or TypeSafe (`TYPESAFE_API_KEY`). Without either,
+`GRAPHWALK_DECISION_FALLBACK=llm` lets any chat model decide (slower, not calibrated).
+
+```python
+import asyncio
+from graphwalk import Index
+
+
+async def main() -> None:
+    async with Index.open("kg.db", escalate_below=0.9) as index:
+        await index.import_triples("kg.nt")  # idempotent
+        result = await index.walk("Where was the director of Inception born?")
+        print(result.best.names, result.confidence, result.escalated, result.cost_usd)
+
+
+asyncio.run(main())
+```
+
+A runnable example over a 15-triple movie graph is in
+[`examples/kg/`](https://github.com/manvirchakal/graphwalk/blob/main/examples/kg/walk.py).
+
+## For agents (MCP)
+
+```bash
+pip install 'graphwalk[mcp]'
+claude mcp add --env OPENROUTER_API_KEY=sk-or-... graphwalk -- graphwalk mcp --db kg.db
+```
+
+Tools: `walk` (answers, paths, confidence), `neighbors`, `get_node`, `status`, plus
+`locate`/`read`/`ingest` for text. The usual loop: call `walk` first; at confidence 0.9
+or above, take the answer; below it, check it with `neighbors` or escalate. The server
+also runs over HTTP with bearer or OAuth 2.1 auth, and ships as a container. See
+[MCP setup](https://manvirchakal.github.io/graphwalk/agents/).
+
+The package carries its own guide for coding agents: `graphwalk guide` (or
+`graphwalk guide --skill` for a `SKILL.md`), and [`llms.txt`](https://github.com/manvirchakal/graphwalk/blob/main/llms.txt).
+
+## Install options
+
+| Extra | Enables |
+|---|---|
+| (none) | import, walk, query, the CLI |
+| `mcp` | the MCP server (stdio and HTTP) |
+| `llm` | LiteLLM: escalation, the LLM decider, text ingestion |
+| `embeddings` | local fastembed embedder (prefilter, dense `locate`) |
+| `eval` | dataset loaders for the benchmarks |
+
+## Documentation
+
+- [Docs site](https://manvirchakal.github.io/graphwalk/): quickstart, agents and MCP,
+  configuration, text ingestion, evidence.
+- [Results](https://github.com/manvirchakal/graphwalk/blob/main/docs/evidence.md): every experiment, with CIs, costs, and the runs behind it.
+- [Design record](https://github.com/manvirchakal/graphwalk/blob/main/docs/design.md) and [roadmap](https://github.com/manvirchakal/graphwalk/blob/main/roadmap.md).
+
+## Status
+
+Alpha (v0.1). The public API is what `graphwalk` exports; submodules may change between
+minor versions. Results are from small samples, mostly one seed. The decision model
+(Jev) is a hosted model from TypeSafe; this project has no affiliation with TypeSafe,
+and the LLM-decider fallback keeps it usable without Jev.
+
+## Development
+
+See [CONTRIBUTING.md](https://github.com/manvirchakal/graphwalk/blob/main/CONTRIBUTING.md).
+
+## License
+
+Apache-2.0. See [LICENSE](https://github.com/manvirchakal/graphwalk/blob/main/LICENSE).

graphwalk-0.1.0/pyproject.toml

@@ -0,0 +1,113 @@
+[project]
+name = "graphwalk"
+version = "0.1.0"
+description = "Cheap, confidence-scored question answering over knowledge graphs: each hop is a calibrated classification decision. Library, CLI, and MCP server."
+readme = "README.md"
+license = "Apache-2.0"
+license-files = ["LICENSE"]
+authors = [{ name = "Manvir Chakal" }]
+keywords = ["knowledge graph", "kgqa", "graph traversal", "mcp", "llm agents", "calibration", "retrieval"]
+requires-python = ">=3.12"
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Intended Audience :: Science/Research",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+    "Typing :: Typed",
+]
+dependencies = [
+    "networkx>=3.4",
+    "numpy>=2.1",
+    "pydantic>=2.12",
+    "pydantic-settings>=2.6",
+    "typer>=0.15",
+    "typesafe-sdk>=0.7.2,<0.8",
+]
+
+[project.optional-dependencies]
+embeddings = [
+    "fastembed>=0.7",
+]
+llm = [
+    "litellm>=1.60",
+]
+mcp = [
+    "mcp>=1.20",
+    "pyjwt[crypto]>=2.10",
+    "uvicorn>=0.34",
+]
+eval = [
+    "httpx>=0.28",
+    "pyarrow>=17",  # FanOutQA's Wikipedia corpus mirror is an Arrow file
+]
+
+[project.urls]
+Homepage = "https://github.com/manvirchakal/graphwalk"
+Documentation = "https://manvirchakal.github.io/graphwalk/"
+Source = "https://github.com/manvirchakal/graphwalk"
+Changelog = "https://github.com/manvirchakal/graphwalk/blob/main/CHANGELOG.md"
+Issues = "https://github.com/manvirchakal/graphwalk/issues"
+
+[project.scripts]
+graphwalk = "graphwalk.cli:app"
+
+[build-system]
+requires = ["uv_build>=0.8.17,<0.9.0"]
+build-backend = "uv_build"
+
+[dependency-groups]
+dev = [
+    "pyright>=1.1.414",
+    "pytest>=9.1.1",
+    "pytest-asyncio>=1.4.0",
+    "pytest-socket>=0.8.1",
+    "ruff>=0.16.9",
+    "types-networkx>=3.6.1.20260911",
+]
+docs = [
+    "mkdocs>=1.6,<2",
+    "mkdocs-material>=9.6",
+]
+
+[tool.ruff]
+line-length = 100
+target-version = "py312"
+src = ["src", "tests"]
+extend-exclude = ["docs"]
+
+[tool.ruff.lint]
+select = [
+    "E", "W", "F", "I", "N", "UP", "B", "A", "C4", "SIM", "RUF",
+    "ASYNC", "PT", "PTH", "TC", "PL", "S", "BLE", "T20", "RET",
+]
+ignore = [
+    "PLR0913", # protocols legitimately take many keyword args
+]
+
+[tool.ruff.lint.per-file-ignores]
+"examples/**" = ["T201", "ASYNC240"]  # small scripts: print output, read files directly
+"tests/**" = ["S101", "S311", "PLR2004", "PLC0415", "ASYNC240"]  # seeded RNGs and blocking fs are fine in tests
+
+[tool.pyright]
+include = ["src", "tests"]
+pythonVersion = "3.12"
+venvPath = "."
+venv = ".venv"
+strict = ["src"]
+extraPaths = ["tests"]
+typeCheckingMode = "standard"
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+pythonpath = ["tests"]
+asyncio_mode = "auto"
+asyncio_default_fixture_loop_scope = "function"
+addopts = ["-ra", "--strict-markers", "--disable-socket", "--allow-unix-socket"]
+markers = [
+    "live: calls a real external API; skipped unless --run-live is given",
+    "neo4j: needs a running Neo4j; skipped unless --run-neo4j is given",
+]

graphwalk-0.1.0/src/graphwalk/__init__.py

@@ -0,0 +1,46 @@
+"""graphwalk: fast, probabilistic question answering over knowledge graphs.
+
+When to use it and how: ``graphwalk.guide()`` (or ``graphwalk guide`` on the command
+line). The public API is what this module exports. Everything else (``graphwalk.traversal``,
+``graphwalk.ingest``, ...) is internal and may change between minor versions.
+"""
+
+from importlib.metadata import version
+
+from graphwalk.core.errors import DocumentNotFoundError, GraphwalkError, StaleLocationError
+from graphwalk.core.model import Edge, Neighbor, Node, Provenance
+from graphwalk.guide import guide
+from graphwalk.index import Index, WalkResult
+from graphwalk.ingest.pipeline import IngestConfig, IngestReport
+from graphwalk.ingest.sources import FileSource, SourceDocument, TextSource
+from graphwalk.ingest.triples import ImportReport
+from graphwalk.locate.documents import FileDocuments, StoredDocuments
+from graphwalk.locate.model import Location, Passage
+from graphwalk.traversal.config import TraversalConfig
+
+__version__ = version("graphwalk")
+
+__all__ = [
+    "DocumentNotFoundError",
+    "Edge",
+    "FileDocuments",
+    "FileSource",
+    "GraphwalkError",
+    "ImportReport",
+    "Index",
+    "IngestConfig",
+    "IngestReport",
+    "Location",
+    "Neighbor",
+    "Node",
+    "Passage",
+    "Provenance",
+    "SourceDocument",
+    "StaleLocationError",
+    "StoredDocuments",
+    "TextSource",
+    "TraversalConfig",
+    "WalkResult",
+    "__version__",
+    "guide",
+]

graphwalk-0.1.0/src/graphwalk/agent_docs/SKILL.md

@@ -0,0 +1,33 @@
+---
+name: graphwalk
+description: Answer questions over an existing knowledge graph (triples, RDF, CSV/JSONL edge lists) by walking it hop by hop with a fast classifier, getting entity answers with a confidence and the path taken; escalate low-confidence queries to an LLM. Use when a user has a knowledge graph and wants cheap, fast multi-hop lookups with a per-answer confidence, and can accept lower accuracy than an LLM writing the query. Not for best-accuracy KG-QA (have an LLM write the query from the relations near the entity), QA over documents (use RAG), or constrained/superlative questions.
+---
+
+# graphwalk
+
+Run `graphwalk guide` (or `python -c "import graphwalk; print(graphwalk.guide())"`)
+for the full guide: when to use it, the measured trade-offs, recipes, and API.
+
+Short version:
+
+1. Check fit first. graphwalk answers from the graph's nodes. It is 6× cheaper and
+   ~4× faster than an LLM writing the query, and gives an informative confidence, but
+   is ~15 F1 points less accurate on curated graphs of any schema size. It loses to
+   multi-step RAG on documents. It does no filtering, ranking, or set intersection.
+2. Import: `graphwalk import kg.nt --graph kg.db` (also `.csv`, `.tsv`, `.jsonl`). No
+   key is needed. Give nodes readable names and types.
+3. Query from Python:
+
+   ```python
+   from graphwalk import Index
+
+   # walks with TraversalConfig.kgqa() by default
+   async with Index.open("kg.db", node_types=[...], escalate_below=0.9) as index:
+       result = await index.walk(question)
+       result.best.names, result.best.path, result.confidence, result.escalated
+   ```
+
+4. Treat `result.confidence` as a ranking signal. Validate the threshold on 50–100
+   labeled questions from the user's own graph before trusting it.
+5. Keys: `OPENROUTER_API_KEY` (Jev decisions; the LLM for escalation). Without Jev,
+   use `GRAPHWALK_DECISION_FALLBACK=llm`; its confidence is not calibrated.