Metadata-Version: 2.4
Name: ai-iq
Version: 5.8.0
Summary: Give your AI long-term memory in 1 command. SQLite-based, no setup, no servers.
Author-email: Kobus Wentzel <kobie@pop.co.za>
License: MIT
Project-URL: Homepage, https://github.com/kobie3717/ai-iq
Project-URL: Repository, https://github.com/kobie3717/ai-iq
Project-URL: Issues, https://github.com/kobie3717/ai-iq/issues
Project-URL: Discord, https://discord.gg/Y2jCXNGgE
Keywords: ai,memory,context,sqlite,rag,semantic-search,knowledge-graph,llm,claude,coding-assistant
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Topic :: Database
Classifier: Topic :: Text Processing :: Indexing
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: dateparser>=1.1.0
Provides-Extra: full
Requires-Dist: numpy>=1.21.0; extra == "full"
Requires-Dist: onnxruntime>=1.14.0; extra == "full"
Requires-Dist: tokenizers>=0.13.0; extra == "full"
Requires-Dist: sqlite-vec>=0.1.0; extra == "full"
Requires-Dist: huggingface-hub>=0.14.0; extra == "full"
Provides-Extra: dev
Requires-Dist: pytest>=7.0; extra == "dev"
Requires-Dist: mypy>=1.0; extra == "dev"
Requires-Dist: pylint>=3.0; extra == "dev"
Dynamic: license-file

# AI-IQ

Give your AI long-term memory in 1 command.

LLMs forget everything. AI-IQ makes them remember.

## Install

```bash
pip install ai-iq
```

## Quick Start

```python
from ai_iq import Memory

memory = Memory()

# Add memories
memory.add("User prefers dark mode", tags=["preference", "ui"])
memory.add("Redis bug fixed with network_mode: host", category="learning")

# Search (hybrid keyword + semantic)
results = memory.search("redis networking")
for r in results:
    print(f"#{r['id']}: {r['content']}")

# Update and delete
memory.update(1, "User STRONGLY prefers dark mode")
memory.delete(1)
```

## CLI

```bash
memory-tool add learning "Docker needs network_mode: host" --project MyApp
memory-tool search "docker networking"
memory-tool dream  # Consolidate duplicates, detect conflicts
```

## Claude Code Plugin

Use AI-IQ directly in Claude Code with auto-capture:

```bash
/plugin marketplace add kobie3717/ai-iq
/plugin install ai-iq
```

See [CLAUDE_CODE_PLUGIN.md](CLAUDE_CODE_PLUGIN.md) for details.

## Why AI-IQ?

- **Single SQLite file = your AI's brain** — No servers, no vector DB, no setup
- **No cloud dependencies** — Works offline, owns your data, zero API keys
- **Works with any Python agent** — Not locked to Claude, OpenAI, or any vendor
- **Hybrid search** — Keyword (FTS5) + semantic (vector) + graph traversal
- **Memories decay naturally** — FSRS-6 algorithm like human memory

## Advanced Features

See [docs/REFERENCE.md](docs/REFERENCE.md) for complete documentation:

- **Passport System** — Complete identity card for any memory (graph connections, provenance chain, access patterns, confidence score)
- **Beliefs & Predictions** — Confidence tracking with Bayesian updates
- **Knowledge Graph** — Entities, relationships, spreading activation
- **Dream Mode** — REM-like consolidation (dedup, conflict detection)
- **Identity Layer** — Auto-discovers behavioral traits
- **Narrative Memory** — Builds cause-effect stories from causal graph
- **Meta-Learning** — Search improves from feedback loops

### Passport System

Every memory has a "passport" — its complete identity card across all dimensions:

```bash
memory-tool passport 42
```

Shows:
- **Core identity**: content, category, project, tags
- **Graph connections**: linked entities with their relationships
- **Memory relationships**: derived-from, related, supersedes chains
- **Provenance**: citations, reasoning, source memories
- **Usage stats**: access count, revisions, FSRS state
- **Passport score**: composite 0-10 score from priority, access patterns, proof count, graph connections, and recency
- **Spreading activation**: related entities discovered via graph traversal

Like a traveler's passport proves who you are and where you've been, a memory passport is its complete dossier.

## Example

See [examples/chatbot_with_memory.py](examples/chatbot_with_memory.py)

## Documentation

[Complete Reference](docs/REFERENCE.md) • [Examples](examples/) • [Architecture](ARCHITECTURE.md)

## Requirements

Python 3.8+ and SQLite 3.37+. Optional: `pip install ai-iq[full]` for semantic search.

## License

MIT

## Links

- **GitHub**: [github.com/kobie3717/ai-iq](https://github.com/kobie3717/ai-iq)
- **PyPI**: [pypi.org/project/ai-iq](https://pypi.org/project/ai-iq/)
- **Discord**: [discord.gg/Y2jCXNGgE](https://discord.gg/Y2jCXNGgE)
