Store, Observe, Learn

Think of it as Supabase for agent context — with built-in skill learning that improves your agents over time.

Multi-modal Storage·Real-time Observation·Skill Learning

Scale from local demos to production without rebuilding context infrastructure — messages, files, skills all in one place with built-in context management.

Cloud Platform

Open source

How It Works

The capabilities that power production AI agents — store context, observe behavior, learn from experience, and monitor everything.

Session

Disk

OpenAI

Platform Capabilities

The production-grade infrastructure your agents need — storage, observability, self-learning, and more.

Context Engineering

Edit, compress, and summarize context on-the-fly — token_limit, middle_out, and session summary strategies keep your agents efficient without modifying stored messages.

Multimodal Context Storage

Unified, persistent storage for all agent data — messages, files, and skills — eliminating fragmented backends (DB, S3, Redis).

Artifact Disk

Filesystem-like workspace to store and share multi-modal outputs (.md, code, reports), ready for multi-agent collaboration.

Background Observer

Automatically extracts tasks from agent conversations and tracks their status in real-time — from pending to running to success or failure.

Self-Learning

Attach sessions to a Learning Space and Acontext automatically distills successful task outcomes into skills — agents improve with every run without manual curation.

SDKs & Integrations

Ready to use with OpenAI, Anthropic, LangGraph, Agno, and other popular agent frameworks.

Acontext Sessions

Store · Retrieve · Multi-Provider Format

Store and retrieve context in OpenAI, Anthropic, or Gemini format with one simple API.

from acontext import AcontextClient
 
client = AcontextClient()
 
# Create a session
session = client.sessions.create()
 
# Store messages (OpenAI format)
client.sessions.store_message(session.id, blob={"role": "user", "content": "Hello!"})
 
# Retrieve in any format: openai, anthropic, gemini
messages = client.sessions.get_messages(session.id, format="anthropic")

# Get token-efficient session summary for prompt injection
summary = client.sessions.get_session_summary(session.id, limit=5)
 
# Apply edit strategies to manage context window size
result = client.sessions.get_messages(
    session.id,
    edit_strategies=[
        {"type": "remove_tool_result", "params": {"keep_recent_n_tool_results": 3}},
        {"type": "token_limit", "params": {"limit_tokens": 30000}}
    ]
)
print(f"Tokens: {result.this_time_tokens}")

Reinvent the Wheel for Each Provider

Focus on creating, not adapting.

1 / 4 · Context Engineering

Join the Community

Connect with early builders & preview new features

Join Discord

Star on GitHub