vestigia 0.1.0__tar.gz

vestigia-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 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
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

vestigia-0.1.0/PKG-INFO

@@ -0,0 +1,379 @@
+Metadata-Version: 2.4
+Name: vestigia
+Version: 0.1.0
+Summary: Evidence attribution for LLM-generated answers using retrieved text chunks.
+Keywords: llm,rag,evidence,attribution,langchain
+Author: Lorenzo Gardini
+Author-email: Lorenzo Gardini <l.gardo98@gmail.com>
+License-Expression: Apache-2.0
+License-File: LICENSE
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Dist: pydantic>=2.14.1
+Requires-Dist: vestigia[typesafe,openai,anthropic] ; extra == 'all'
+Requires-Dist: langchain-anthropic>=1.7.7 ; extra == 'anthropic'
+Requires-Dist: langchain-openai>=1.7.3 ; extra == 'openai'
+Requires-Dist: langchain-typesafe>=0.0.1a3 ; extra == 'typesafe'
+Requires-Dist: pyyaml>=6.0 ; extra == 'typesafe'
+Requires-Python: >=3.12
+Project-URL: Homepage, https://github.com/Lorenzo-Gardini/vestigia
+Project-URL: Repository, https://github.com/Lorenzo-Gardini/vestigia
+Project-URL: Issues, https://github.com/Lorenzo-Gardini/vestigia/issues
+Provides-Extra: all
+Provides-Extra: anthropic
+Provides-Extra: openai
+Provides-Extra: typesafe
+Description-Content-Type: text/markdown
+
+<div align="center">
+
+<img src="https://raw.githubusercontent.com/Lorenzo-Gardini/vestigia/main/assets/vestigia-banner.svg" alt="Vestigia - Trace answers back to evidence. An illustrated magnifying glass connects source passages to an answer." width="100%">
+
+Classify how retrieved text supports, partially supports, contradicts, or does
+not contribute to an answer.
+
+<p>
+  <a href="https://github.com/Lorenzo-Gardini/vestigia/actions/workflows/tests.yml">
+    <img src="https://github.com/Lorenzo-Gardini/vestigia/actions/workflows/tests.yml/badge.svg?branch=main" alt="Tests">
+  </a>
+  <a href="https://pypi.org/project/vestigia/">
+    <img src="https://img.shields.io/pypi/v/vestigia" alt="PyPI version">
+  </a>
+  <img src="https://img.shields.io/badge/python-3.12%2B-blue" alt="Python 3.12 or newer">
+</p>
+
+<p>
+  <a href="#features">Features</a> ·
+  <a href="#quick-start">Quick start</a> ·
+  <a href="#classification-statuses">Statuses</a> ·
+  <a href="#development">Development</a> ·
+  <a href="#releasing-to-pypi">Releases</a>
+</p>
+
+</div>
+
+## Features
+
+- **Two classification approaches:** structured chat-model output and a
+  TypeSafe system-one classifier.
+- **Traceable results:** each result retains the original chunk and its stable
+  identifier.
+- **Typed data:** Pydantic models with frozen attributes validate inputs and outputs.
+- **Packaged prompts:** Markdown and YAML prompt templates are available through
+  Python APIs.
+- **Optional integrations:** install only the provider dependencies you need.
+
+## Quick start
+
+Requires Python 3.12 or newer. Install the OpenAI integration with your package
+manager:
+
+```bash
+# pip (inside your virtual environment)
+python -m pip install "vestigia[openai]"
+
+# uv (in a uv-managed project)
+uv add "vestigia[openai]"
+
+# Poetry (in a Poetry-managed project)
+poetry add "vestigia[openai]"
+```
+
+Set `OPENAI_API_KEY` in your environment. Save the following example as
+`classify.py` and run it with `python classify.py` (or `uv run classify.py`,
+`poetry run python classify.py`).
+This example sends data to the configured provider and may incur API charges.
+
+```python
+import asyncio
+
+from langchain_openai import ChatOpenAI
+
+from vestigia.chunk_classification import Chunk
+from vestigia.chunk_classification.langchain import ChatModelChunkClassifier
+
+classifier = ChatModelChunkClassifier(
+    ChatOpenAI(model="gpt-4o-mini"),
+)
+
+
+async def classify_answer():
+    result = await classifier.classify(
+        query="What is the repayment term?",
+        answer="The plan lasts 60 months.",
+        chunks=[
+            Chunk(
+                id="repayment-term",
+                text="The repayment plan lasts 60 months.",
+            ),
+            Chunk(
+                id="early-repayment",
+                text="Early repayment has no additional fee.",
+            ),
+        ],
+    )
+
+    for item in result.classifications:
+        print(item.chunk.id, item.status.value, item.explanation)
+
+
+if __name__ == "__main__":
+    asyncio.run(classify_answer())
+```
+
+`ChatModelChunkClassifier` accepts any LangChain `BaseChatModel` that supports
+structured output; the OpenAI integration above is one option.
+
+### Use a TypeSafe system-one model
+
+Install the extra with `python -m pip install "vestigia[typesafe]"`
+(`uv add` or `poetry add` work the same way) and set `TYPESAFE_API_KEY`.
+Save this example as `classify.py` and run it with
+`python classify.py`:
+
+```python
+import asyncio
+
+from langchain_typesafe import TypeSafeClassifier
+
+from vestigia.chunk_classification import Chunk
+from vestigia.chunk_classification.langchain import SystemOneModelChunkClassifier
+
+
+async def classify_answer():
+    classifier = SystemOneModelChunkClassifier(TypeSafeClassifier())
+    result = await classifier.classify(
+        query="What is the repayment term?",
+        answer="The plan lasts 60 months.",
+        chunks=[Chunk(id="source-1", text="The plan lasts 60 months.")],
+    )
+    for item in result.classifications:
+        print(item.chunk.id, item.status.value, item.probabilities, item.confidence)
+
+
+if __name__ == "__main__":
+    asyncio.run(classify_answer())
+```
+
+Like the chat-model example, this makes a real provider request. System-one
+results include per-status probabilities and confidence.
+
+### Use the data models directly
+
+The package exports its domain models from `vestigia.chunk_classification`:
+
+```python
+from vestigia.chunk_classification import (
+    ChatModelChunkClassification,
+    ChatModelChunksClassificationResult,
+    Chunk,
+    ChunkStatus,
+)
+
+chunk = Chunk(id="source-1", text="The plan lasts 60 months.")
+result = ChatModelChunksClassificationResult(
+    classifications=(
+        ChatModelChunkClassification(
+            chunk=chunk,
+            status=ChunkStatus.USED,
+            explanation="The answer states the same repayment term.",
+        ),
+    )
+)
+```
+
+## Classification statuses
+
+| Status | Meaning |
+| --- | --- |
+| `used` | The answer states or faithfully paraphrases meaningful information from the chunk. |
+| `partially_used` | The answer uses meaningful information but omits or incompletely represents other significant details. |
+| `contradicted` | The answer makes a claim incompatible with information in the chunk. |
+| `not_used` | No meaningful information from the chunk appears in the answer, and it is not contradicted. |
+| `uncertain` | The available text does not support a reliable classification. |
+
+Omission alone is not a contradiction. Both packaged prompts instruct the model
+to assess each chunk against the answer and avoid relying on external knowledge.
+The labels describe observable overlap and conflict, not proof of the generating
+model's internal reasoning or actual use of a retrieved chunk.
+
+## Classifiers and prompts
+
+| API | Purpose | Optional dependency |
+| --- | --- | --- |
+| `ChatModelChunkClassifier` | Classify chunks using a LangChain chat model with structured output. | `openai`, `anthropic`, or another installed LangChain chat integration |
+| `SystemOneModelChunkClassifier` | Classify chunks using a TypeSafe system-one model and per-status criteria. | `typesafe` |
+
+The adapters are exported from `vestigia.chunk_classification.langchain`.
+`ChunkClassifier` is exported from `vestigia.chunk_classification.classifier`.
+
+Load the packaged prompt templates directly when needed:
+
+```python
+from vestigia.chunk_classification.prompts import (
+    load_chat_model_prompt,
+    load_system_one_prompt,
+)
+
+chat_prompt_text = load_chat_model_prompt()
+system_one_template = load_system_one_prompt()
+```
+
+`load_system_one_prompt()` requires the `typesafe` extra, which includes
+PyYAML.
+
+### Custom prompts
+
+Pass a `langchain_core.prompts.PromptTemplate` as `prompt=` to the chat
+classifier. Its available variables are `query`, `answer`, and `chunks`; the
+last contains XML-escaped source text and identifiers. The model returns only
+each chunk's `chunk_id`, `status`, and optional `explanation`; results reference
+the original `Chunk` objects.
+
+For the system-one classifier, pass a `SystemOneModelTemplate`:
+
+```python
+from vestigia.chunk_classification import ChunkStatus, SystemOneModelTemplate
+
+template = SystemOneModelTemplate(
+    state="Question: {query}\nAnswer: {answer}",
+    instruction="Evaluate the following evidence: {chunk}",
+    criteria={
+        ChunkStatus.USED: "Meaningful information appears in the answer.",
+        ChunkStatus.NOT_USED: "No meaningful information appears in the answer.",
+    },
+)
+```
+
+The state template receives `query` and `answer`; the instruction template
+receives each chunk's text as `chunk`. Criteria require at least two statuses.
+
+## Input and output contracts
+
+- Chunk IDs must be unique within each call. Duplicates raise `ValueError`
+  before a provider is invoked.
+- An empty chunk sequence returns `classifications=()` without a provider call.
+- Results follow input order and contain the original `Chunk` objects. Missing,
+  extra, or duplicate response IDs raise `ValueError`.
+- System-one answers must be choice answers whose labels belong to the prompt's
+  criteria. Confidence and probability values must be finite and in `[0, 1]`.
+  The package does not normalize the distribution or require an exact sum of 1.
+- Backend, prompt-formatting, and Pydantic validation errors propagate to the
+  caller; the package does not substitute partial or empty success responses.
+- Model attributes are frozen and extra fields are rejected. Freezing is
+  shallow: nested dictionaries remain mutable, and normal Pydantic type
+  coercion applies.
+
+## Installation options
+
+The PyPI distribution, Python import package, and GitHub repository are all
+named `vestigia`.
+
+Install the core models without provider integrations:
+
+```bash
+python -m pip install vestigia
+```
+
+Install optional integrations with `python -m pip install "vestigia[<extra>]"`,
+`uv add "vestigia[<extra>]"`, or `poetry add "vestigia[<extra>]"`:
+
+| Extra | Includes |
+| --- | --- |
+| `openai` | LangChain OpenAI integration |
+| `typesafe` | TypeSafe classifier and YAML prompt support |
+| `anthropic` | LangChain Anthropic integration |
+| `all` | All optional integrations |
+
+`ChatModelChunkClassifier` accepts any LangChain `BaseChatModel`, so the
+`openai` and `anthropic` extras are conveniences rather than requirements.
+
+## Development
+
+Install [uv](https://docs.astral.sh/uv/), clone the repository, and install the
+test and classifier dependencies:
+
+```bash
+git clone https://github.com/Lorenzo-Gardini/vestigia.git
+cd vestigia
+uv sync --locked --group dev --all-extras
+```
+
+Run the test suite:
+
+```bash
+uv run --locked --group dev --all-extras pytest
+```
+
+The tests use real LangChain runnables and TypeSafe response models with mocked
+providers. They require neither API keys nor network access to a model.
+Documentation examples are tested without making real provider requests.
+GitHub Actions runs on pushes and pull requests, checking the base installation,
+each classifier extra separately, and the combined configuration on Python 3.12.
+It also builds distributions, validates their metadata and README, and checks
+imports and packaged resources from an isolated wheel installation.
+
+For architecture and contribution guidance, see [AGENTS.md](AGENTS.md).
+
+## Releasing to PyPI
+
+Publishing uses [PyPI Trusted Publishing](https://docs.pypi.org/trusted-publishers/):
+no PyPI API token needs to be stored in GitHub.
+
+### Publishing configuration
+
+Releases upload through the `pypi` GitHub environment in this repository's
+settings; configure required reviewers there for manual approval before an
+upload. The [vestigia PyPI project](https://pypi.org/manage/project/vestigia/settings/publishing/)
+trusts this GitHub publisher:
+
+| Setting | Value |
+| --- | --- |
+| PyPI project name | `vestigia` |
+| Owner | `Lorenzo-Gardini` |
+| Repository | `vestigia` |
+| Workflow filename | `publish.yml` |
+| Environment | `pypi` |
+
+Ensure the PyPI Trusted Publisher configuration uses the current repository,
+workflow, and environment values shown above.
+
+### Publish a version
+
+1. Update `project.version` in [pyproject.toml](pyproject.toml), run `uv lock`,
+   and commit the version and lockfile together.
+2. Verify the tests and packaging locally:
+
+   ```bash
+   uv run --locked --group dev --all-extras pytest
+   uv build --no-sources
+   uvx --from twine twine check --strict dist/*
+   ```
+
+   Use a clean `dist/` directory so old artifacts are not included.
+3. Merge the changes into the default branch and push them before releasing.
+4. Publish a GitHub Release targeting that commit, with a tag exactly matching
+   `v<project.version>` (for example, `v0.1.0`). Creating a tag alone or saving a
+   draft release does not publish to PyPI.
+5. The [publishing workflow](.github/workflows/publish.yml) checks the version,
+   runs all CI profiles, builds and validates the packages, then uploads those
+   same artifacts to PyPI after any environment approval.
+6. Confirm the new version on [PyPI](https://pypi.org/project/vestigia/) and
+   install it in a fresh virtual environment:
+
+   ```bash
+   python -m pip install "vestigia[openai]==0.1.0"
+   ```
+
+   Replace `0.1.0` with the version being verified.
+
+PyPI releases cannot be overwritten. For a corrected artifact, bump the version
+and publish a new release.
+
+## License
+
+Licensed under [Apache-2.0](LICENSE).