Metadata-Version: 2.4
Name: agentmem-core
Version: 0.1.0
Summary: A proactive memory layer for long-horizon coding agents, it knows when to remind.
Project-URL: Homepage, https://agentmem.xyz
Project-URL: Repository, https://github.com/agentmem/agentmem
Project-URL: Documentation, https://agentmem.xyz/docs
Project-URL: Changelog, https://github.com/agentmem/agentmem/blob/main/CHANGELOG.md
Author: The AgentMem Authors
License-Expression: Apache-2.0
Keywords: agents,claude-code,coding-agents,llm,long-horizon,memory
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Typing :: Typed
Requires-Python: >=3.11
Requires-Dist: anthropic>=0.40
Requires-Dist: pydantic-settings>=2.3
Requires-Dist: pydantic>=2.7
Provides-Extra: agent-sdk
Requires-Dist: claude-agent-sdk>=0.2; extra == 'agent-sdk'
Provides-Extra: aider
Requires-Dist: aider-chat<0.87,>=0.86; extra == 'aider'
Provides-Extra: daemon
Requires-Dist: agentmem-daemon; extra == 'daemon'
Provides-Extra: litellm
Requires-Dist: litellm>=1.40; extra == 'litellm'
Provides-Extra: mcp
Requires-Dist: mcp<2,>=1.28; extra == 'mcp'
Provides-Extra: openai-agents
Requires-Dist: openai-agents<0.19,>=0.18; extra == 'openai-agents'
Description-Content-Type: text/markdown

# agentmem-core

The core of [AgentMem](https://github.com/agentmem/agentmem): a proactive memory layer
for long-horizon coding agents. It maintains a structured memory bank and decides *when*
to remind your agent, instead of dumping everything or retrieving on similarity.

```bash
pip install agentmem-core
```

Published as `agentmem-core` (the name `agentmem` was already taken on PyPI by an
unrelated project); everything you import and run is still `agentmem`.

```python
from agentmem import MemorySession, triggers

mem = MemorySession(task="Fix the auth tests without changing the public API")
while not done:
    reminder = mem.pending_context()          # str | None
    reply = call_your_agent(messages, memory_context=reminder)
    mem.observe(reply.new_messages)
```

Full docs, integrations, and the design rationale live in the
[main repository](https://github.com/agentmem/agentmem).

Licensed under Apache-2.0.
