letta

Letta framework for building stateful AI agents with long-term memory. Use for AI agent development, memory management, tool integration, and multi-agent systems.

⭐4stars🔀0forks📁View on GitHub🕐Updated Dec 29, 2025⬇️1downloads

When & Why to Use This Skill

The Letta skill provides comprehensive support for building stateful AI agents using the Letta framework. It enables developers to implement long-term memory, manage complex agent states, and integrate tools seamlessly into multi-agent systems, ensuring persistent and context-aware AI interactions. This skill is essential for developers looking to move beyond stateless LLM calls to sophisticated, memory-retaining AI architectures.

Use Cases

  • Case 1: Developing AI agents that require persistent long-term memory to maintain user context and history across multiple sessions.
  • Case 2: Implementing and debugging stateful agent architectures using the Letta Python SDK or REST API.
  • Case 3: Managing complex tool integrations and 'blocks' within the Letta ecosystem to extend agent capabilities.
  • Case 4: Building and orchestrating multi-agent systems that require shared state and synchronized memory management.
  • Case 5: Referencing official Letta documentation for best practices in agent deployment and memory optimization.
nameletta
descriptionLetta framework for building stateful AI agents with long-term memory. Use for AI agent development, memory management, tool integration, and multi-agent systems.

Letta Skill

Comprehensive assistance with letta development, generated from official documentation.

When to Use This Skill

This skill should be triggered when:

  • Working with letta
  • Asking about letta features or APIs
  • Implementing letta solutions
  • Debugging letta code
  • Learning letta best practices

Quick Reference

Common Patterns

Quick reference patterns will be added as you use the skill.

Example Code Patterns

Example 1 (bash):

pip install letta-evals

Example 2 (bash):

uv pip install letta-evals

Example 3 (python):

import requests

url = "https://api.letta.com/v1/mcp-servers/mcp_server_id"

headers = {"Authorization": "Bearer <token>"}

response = requests.delete(url, headers=headers)

print(response.json())

Example 4 (python):

from letta_client import Letta

client = Letta(
    project="YOUR_PROJECT",
    token="YOUR_TOKEN",
)
client.groups.delete(
    group_id="group-123e4567-e89b-42d3-8456-426614174000",
)

Reference Files

This skill includes comprehensive documentation in references/:

  • agents.md - Agents documentation
  • api_sdk.md - Api Sdk documentation
  • blocks.md - Blocks documentation
  • deployment.md - Deployment documentation
  • getting_started.md - Getting Started documentation
  • messages.md - Messages documentation
  • multiprocessing.md - Multiprocessing documentation
  • other.md - Other documentation
  • tools.md - Tools documentation

Use view to read specific reference files when detailed information is needed.

Working with This Skill

For Beginners

Start with the getting_started or tutorials reference files for foundational concepts.

For Specific Features

Use the appropriate category reference file (api, guides, etc.) for detailed information.

For Code Examples

The quick reference section above contains common patterns extracted from the official docs.

Resources

references/

Organized documentation extracted from official sources. These files contain:

  • Detailed explanations
  • Code examples with language annotations
  • Links to original documentation
  • Table of contents for quick navigation

scripts/

Add helper scripts here for common automation tasks.

assets/

Add templates, boilerplate, or example projects here.

Notes

  • This skill was automatically generated from official documentation
  • Reference files preserve the structure and examples from source docs
  • Code examples include language detection for better syntax highlighting
  • Quick reference patterns are extracted from common usage examples in the docs

Updating

To refresh this skill with updated documentation:

  1. Re-run the scraper with the same configuration
  2. The skill will be rebuilt with the latest information