Metadata-Version: 2.4
Name: memgentic
Version: 0.8.0
Summary: Universal AI memory layer — cross-tool knowledge capture with MCP, semantic search, and knowledge graphs
Project-URL: Homepage, https://github.com/Chariton-kyp/Memgentic
Project-URL: Repository, https://github.com/Chariton-kyp/Memgentic
Project-URL: Documentation, https://github.com/Chariton-kyp/Memgentic/tree/main/docs
Project-URL: Issues, https://github.com/Chariton-kyp/Memgentic/issues
Project-URL: Changelog, https://github.com/Chariton-kyp/Memgentic/blob/main/CHANGELOG.md
Project-URL: Source, https://github.com/Chariton-kyp/Memgentic/tree/main/memgentic
Author-email: Chariton Kypraios <chariton@ellinai.com>
License-Expression: Apache-2.0
License-File: LICENSE
Keywords: agents,ai,chatgpt,claude,codex,cursor,gemini,knowledge-graph,llm,mcp,memory,rag,semantic-search,vector-database
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Information Technology
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Database
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Typing :: Typed
Requires-Python: >=3.12
Requires-Dist: aiosqlite>=0.22
Requires-Dist: click>=8.1
Requires-Dist: httpx>=0.28
Requires-Dist: mcp[cli]>=1.26
Requires-Dist: pydantic-settings>=2.10
Requires-Dist: pydantic>=2.12
Requires-Dist: python-dotenv>=1.0
Requires-Dist: pyyaml>=6.0
Requires-Dist: qdrant-client>=1.17
Requires-Dist: rich>=14.0
Requires-Dist: sqlite-vec>=0.1.9
Requires-Dist: structlog>=25.0
Requires-Dist: tenacity>=9.0
Requires-Dist: watchdog>=6.0
Provides-Extra: dev
Requires-Dist: google-genai>=1.50; extra == 'dev'
Requires-Dist: langchain-anthropic>=1.0; extra == 'dev'
Requires-Dist: langchain-core>=1.2.28; extra == 'dev'
Requires-Dist: langchain-google-genai>=4.0; extra == 'dev'
Requires-Dist: langchain-ollama>=0.3; extra == 'dev'
Requires-Dist: langgraph>=1.1; extra == 'dev'
Requires-Dist: networkx>=3.4; extra == 'dev'
Requires-Dist: pyright>=1.1; extra == 'dev'
Requires-Dist: pytest-asyncio>=1.0; extra == 'dev'
Requires-Dist: pytest-cov>=6.0; extra == 'dev'
Requires-Dist: pytest-timeout>=2.3; extra == 'dev'
Requires-Dist: pytest>=9.0.3; extra == 'dev'
Requires-Dist: rapidfuzz>=3.9; extra == 'dev'
Requires-Dist: ruff>=0.14; extra == 'dev'
Provides-Extra: documents
Requires-Dist: ebooklib>=0.18; extra == 'documents'
Requires-Dist: openpyxl>=3.1; extra == 'documents'
Requires-Dist: pdfminer-six>=20231228; extra == 'documents'
Requires-Dist: python-docx>=1.1; extra == 'documents'
Requires-Dist: python-pptx>=1.0; extra == 'documents'
Requires-Dist: striprtf>=0.0.26; extra == 'documents'
Provides-Extra: intelligence
Requires-Dist: google-genai>=1.50; extra == 'intelligence'
Requires-Dist: langchain-anthropic>=1.0; extra == 'intelligence'
Requires-Dist: langchain-core>=1.2.28; extra == 'intelligence'
Requires-Dist: langchain-google-genai>=4.0; extra == 'intelligence'
Requires-Dist: langchain-ollama>=0.3; extra == 'intelligence'
Requires-Dist: langgraph>=1.1; extra == 'intelligence'
Requires-Dist: networkx>=3.4; extra == 'intelligence'
Requires-Dist: rapidfuzz>=3.9; extra == 'intelligence'
Provides-Extra: observability
Requires-Dist: opentelemetry-api>=1.28; extra == 'observability'
Requires-Dist: opentelemetry-exporter-otlp>=1.28; extra == 'observability'
Requires-Dist: opentelemetry-instrumentation-httpx>=0.49; extra == 'observability'
Requires-Dist: opentelemetry-instrumentation-sqlite3>=0.49; extra == 'observability'
Requires-Dist: opentelemetry-sdk>=1.28; extra == 'observability'
Provides-Extra: reranker
Requires-Dist: llama-cpp-python>=0.3.5; extra == 'reranker'
Provides-Extra: sqlite-vec
Description-Content-Type: text/markdown

# memgentic

**Universal AI memory layer** — zero-effort knowledge capture across every AI tool you use. Source-aware memory with semantic search, filtering, and knowledge graphs.

`memgentic` is the **core engine**. Install it to get the CLI, the MCP server, the ingestion pipeline, and the storage layer.

```bash
pip install memgentic
# or with the REST API:
pip install 'memgentic-api'
# or with native Rust acceleration:
pip install 'memgentic[native]'
```

## What it does

Every AI conversation is ephemeral — knowledge is lost when the session ends. Worse, it's siloed: what Claude knows, ChatGPT doesn't. `memgentic` captures knowledge automatically from every AI tool you use and makes it available everywhere via MCP.

- **Local-first.** Data stays on your machine. Works offline.
- **Source-aware.** Every memory carries provenance: which tool, which session, which timestamp.
- **Semantic search.** Qwen3-Embedding-0.6B via Ollama, 768-dim, Apache 2.0.
- **Recall Tiers (T0–T4).** Progressive context loader — persona, horizon, orbit, deep, atlas.
- **Watchers.** Auto-capture from Claude Code, Codex CLI, Gemini CLI, Copilot CLI, Aider, Antigravity, ChatGPT/Claude web imports, Cursor, Windsurf.
- **Chronograph.** Bitemporal knowledge graph — entities, relationships, validity windows.
- **27 MCP tools.** Full surface exposed via the MCP protocol for any compatible client.

## Quick start

```bash
# One-shot setup (detects installed AI tools, configures models, installs hooks)
memgentic init

# Run the MCP server + the file watcher in one process
memgentic serve --watch

# Semantic search from the CLI
memgentic search "react performance tips"

# Manual store
memgentic remember "Fixed JWT expiry bug: iat claim was missing"
```

See the [project README on GitHub](https://github.com/Chariton-kyp/Memgentic) for architecture, full feature matrix, MCP tool reference, and benchmarks.

## License

Apache 2.0. See [LICENSE](https://github.com/Chariton-kyp/Memgentic/blob/main/LICENSE).
