Metadata-Version: 2.4
Name: nexus-agentos
Version: 1.1.7
Summary: Agent Operating System — v1.1.7: DAG Checkpoint & Restore for long-running tool workflows
Author: AgentOS Team
License: MIT
Keywords: agent,ai,checkpoint,framework,llm,multi-agent,orchestration
Classifier: Development Status :: 6 - Mature
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.11
Requires-Dist: aiosqlite>=0.20.0
Requires-Dist: fastapi>=0.110.0
Requires-Dist: httpx>=0.27.0
Requires-Dist: langsmith>=0.1.0
Requires-Dist: numpy>=1.26.0
Requires-Dist: openai>=1.0.0
Requires-Dist: pydantic>=2.0
Requires-Dist: pyyaml>=6.0
Requires-Dist: uvicorn[standard]>=0.29.0
Provides-Extra: full
Requires-Dist: chromadb>=0.5.0; extra == 'full'
Requires-Dist: faiss-cpu>=1.8.0; extra == 'full'
Requires-Dist: sentence-transformers>=3.0.0; extra == 'full'
Description-Content-Type: text/markdown

# Nexus AgentOS v1.1.7

Production-ready multi-agent framework with **Checkpoint/Recovery** for long-running DAG workflows.

## v1.1.7 New: Checkpoint & Restore

ToolOrchestrator now supports DAG execution checkpointing:

```python
from agentos import ToolOrchestrator, DAGBuilder, CheckpointData, chain_builder

orch = ToolOrchestrator(registry)
dag = chain_builder("my_chain", ["tool_a", "tool_b", "tool_c"])

# Save checkpoint mid-execution
cp = orch.checkpoint(dag)  # → CheckpointData
cp.to_json()  # serialize to disk/DB

# Later: restore and resume from checkpoint
orch.restore_from_checkpoint(dag, checkpoint)
await orch.execute(dag)  # skips completed nodes, resumes from breakpoint

# Auto-checkpointing
await orch.execute_with_checkpoint(dag, checkpoint_callback=save_to_db)
```

## Installation

```bash
pip install nexus-agentos==1.1.7
```
