Metadata-Version: 2.5
Name: palimem
Version: 0.0.1
Summary: Justified memory for LLM agents: evidence-pinned beliefs, retraction that propagates, uncertainty kept rather than guessed. Pre-alpha.
Project-URL: Repository, https://github.com/chleiva/palimem
Project-URL: Study, https://doi.org/10.5281/zenodo.23127764
Author: Christian Leiva Beltran
License: MIT
License-File: LICENSE
License-File: LICENSE-CC-BY-4.0
Keywords: agents,belief-revision,llm,memory,truth-maintenance
Classifier: Development Status :: 2 - Pre-Alpha
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.11
Provides-Extra: anthropic
Requires-Dist: anthropic>=0.40; extra == 'anthropic'
Provides-Extra: bedrock
Requires-Dist: boto3>=1.34; extra == 'bedrock'
Provides-Extra: dev
Requires-Dist: jsonschema>=4; extra == 'dev'
Requires-Dist: pytest>=8; extra == 'dev'
Requires-Dist: ruff>=0.5; extra == 'dev'
Provides-Extra: mcp
Requires-Dist: mcp>=1.0; extra == 'mcp'
Provides-Extra: openai-compat
Requires-Dist: openai>=1.0; extra == 'openai-compat'
Description-Content-Type: text/markdown

# palimem

**Justified memory for LLM agents.** Beliefs are held with the evidence that justifies them; withdrawing or correcting a report repairs every conclusion that depended on it; unresolved alternatives are kept rather than guessed.

> **Status: pre-alpha (0.0.x). Nothing here works yet.** This package name is reserved while the SDK is built in the open. Contracts may break at any 0.x release; 1.0 means the gate-G2 acceptance suite passes.

palimem is the SDK built on the revision kernel validated in the PALIMPSEST study
([DOI 10.5281/zenodo.23127764](https://doi.org/10.5281/zenodo.23127764), code at `chleiva/palimpsest`).

- Plan and review: [`docs/PROPOSAL.md`](docs/PROPOSAL.md)
- Task tracker: [`docs/TASKS.md`](docs/TASKS.md)

Licence: MIT (code), CC BY 4.0 (documentation and data).
