Metadata-Version: 2.4
Name: episodial
Version: 0.1.1
Summary: Python SDK for the Episodial cognitive memory API
Project-URL: Homepage, https://github.com/episodial/sdk-python
Project-URL: Documentation, https://docs.episodial.ai
Author: Episodial
License-Expression: MIT
Keywords: agents,ai,cognitive,episodial,llm,memory
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Typing :: Typed
Requires-Python: >=3.9
Requires-Dist: httpx>=0.25.0
Provides-Extra: dev
Requires-Dist: mypy>=1.0; extra == 'dev'
Requires-Dist: pytest-asyncio>=0.21; extra == 'dev'
Requires-Dist: pytest>=7.0; extra == 'dev'
Requires-Dist: respx>=0.21; extra == 'dev'
Description-Content-Type: text/markdown

# episodial

Python SDK for the Episodial cognitive memory API. Includes both synchronous and async clients.

## Install

```bash
pip install episodial
```

## Quick Start

```python
from episodial import Episodial

ep = Episodial(
    base_url="https://api.episodial.ai",
    api_key="ep_live_...",
    project_id="my-app",
)

# Store a memory
ep.remember(
    user_id="user_123",
    user_message="How do I reset my password?",
    assistant_message="Go to Settings > Security > Reset password.",
)

# Recall relevant memories
result = ep.recall("user_123", "password help")
for mem in result.memories:
    print(mem.content, mem.salience)
```

## Async Client

```python
import asyncio
from episodial import AsyncEpisodial

async def main():
    async with AsyncEpisodial(
        base_url="https://api.episodial.ai",
        api_key="ep_live_...",
    ) as ep:
        result = await ep.recall("user_123", "password help")
        print(result.memories)

asyncio.run(main())
```

## Configuration

```python
ep = Episodial(
    base_url="http://localhost:8000",  # Memory Core URL
    api_key="ep_live_...",             # API key (optional in dev)
    organization_id="org_...",         # For billing endpoints
    project_id="default",             # Default project for all calls
    timeout=30.0,                     # Request timeout in seconds
)
```

## Memory Operations

### Ingest

```python
ep.ingest(
    user_id="user_123",
    project_id="my-app",
    chat_id="conv_1",        # optional — scope to a conversation
    event_type="llm_interaction",
    payload={
        "user_message": "What is Kubernetes?",
        "assistant_message": "Kubernetes is a container orchestration platform...",
    },
    initial_salience=0.7,
)
```

Or use the convenience method:

```python
ep.remember(
    user_id="user_123",
    user_message="What is Kubernetes?",
    assistant_message="Kubernetes is a container orchestration platform...",
    chat_id="conv_1",        # optional — scope to a conversation
    salience=0.7,
)
```

### Retrieve

```python
result = ep.retrieve(
    user_id="user_123",
    project_id="my-app",
    chat_id="conv_1",        # optional — filter by conversation
    task_goal="Help user with deployment",
    token_budget=2000,
)

for memory in result.memories:
    print(memory.content, memory.memory_type, memory.salience)
```

Or use the shorthand:

```python
result = ep.recall("user_123", "deployment help", chat_id="conv_1")
```

### Conversation Scoping

Use `chat_id` to isolate memories per conversation. Omit it (or pass `None`) for
cross-conversation global memories.

```python
# Retrieve memories from this chat + global memories (default)
result = ep.retrieve(
    user_id="user_123",
    project_id="my-app",
    chat_id="conv_1",
    include_global=True,
    task_goal="deployment help",
)

# Only memories from this specific chat
result = ep.retrieve(
    user_id="user_123",
    project_id="my-app",
    chat_id="conv_1",
    chat_only=True,
    task_goal="deployment help",
)

# All memories for the user (no chat filter)
result = ep.retrieve(
    user_id="user_123",
    project_id="my-app",
    task_goal="deployment help",
)
```

### Feedback

```python
ep.helpful("memory_id_123")    # positive feedback
ep.unhelpful("memory_id_456")  # negative feedback

# Fine-grained control
ep.feedback(
    memory_id="memory_id_123",
    feedback_type="success",
    value=0.8,
    context="User confirmed this solved their issue",
)
```

### Forget

```python
# Forget specific memories
ep.forget(
    memory_ids=["mem_1", "mem_2"],
    reason="User requested data deletion",
)

# Forget all memories for a user
ep.forget(
    user_id="user_123",
    project_id="my-app",
    reason="Account deletion",
)
```

### Consolidation

```python
ep.consolidate("user_123", "my-app")
```

### Procedural Learning

```python
ep.mine_procedures("user_123", "my-app")

result = ep.match_procedures(
    "user_123",
    "user wants to deploy to production",
)
for proc in result.procedures:
    print(proc.readable, proc.confidence)
```

## Billing

```python
balance = ep.get_balance()
print(f"${balance.balance_dollars} remaining")

pricing = ep.get_pricing()

ep.topup("growth")  # $50 + 20% bonus
```

## Error Handling

```python
from episodial import Episodial, EpisodialError, InsufficientBalanceError

try:
    ep.recall("user_123", "hello")
except InsufficientBalanceError:
    print("Balance too low — top up!")
except EpisodialError as e:
    print(f"API error {e.status}: {e}")
```

## Context Manager

Both clients support context managers for automatic cleanup:

```python
with Episodial(api_key="ep_live_...") as ep:
    ep.remember(
        user_id="u1",
        user_message="hi",
        assistant_message="hello!",
    )
# httpx client closed automatically

async with AsyncEpisodial(api_key="ep_live_...") as ep:
    await ep.remember(
        user_id="u1",
        user_message="hi",
        assistant_message="hello!",
    )
```

## License

MIT
