llm-kb
← Back to products
Productlaunched

Self-hosted shared memory for AI agents with policy-controlled summaries

Luthn is an open-source, self-hosted shared memory space for AI agents. It keeps raw documents and sensitive records behind explicit boundaries, providing agents with only policy-approved summaries, references, and context packs. This solves the problem of agents needing shared context while avoiding privacy and access risks from copying raw data into every session.

Visit project ↗
1 engagement·1 source·Mon, Jul 13, 2026, 03:21 AM
Runs on infrastructure you control. Includes usage history. Supports HTTP access. Open-sourced on GitHub. Looking for early users.

Related

Product

Collaborative context-sharing memory platform for AI agents and teams

A platform that enables AI agents and human teams to share and persist context across sessions. It uses LLMs to manage memory, allowing agents to recall past interactions and collaborate on tasks. Solves the problem of fragmented context in multi-agent systems and team workflows.

2 engagement·1 source·hackernews
Sun, Jul 12, 2026, 10:28 PM
Product

Attestor: open-source zero-trust boundary for autonomous AI agents

Attestor is an open-source execution boundary that enforces zero-trust security for autonomous AI agents. It uses LLMs to define and enforce policies that restrict agent actions, preventing unauthorized access or data leaks. This solves the problem of safely deploying autonomous agents in production environments.

3 engagement·1 source·hackernews
Sun, Jul 12, 2026, 07:54 AM
Product

Runeward: Sandbox AI agents with policy gates

Runeward is a sandboxing tool for AI agents that enforces policy gates to restrict agent actions. It uses LLMs to interpret and enforce user-defined policies, solving the problem of unsafe or unintended agent behavior for developers building autonomous AI systems.

1 engagement·1 source·hackernews
Sun, Jul 12, 2026, 09:35 AM
Paper

Shared Selective Persistent Memory Architecture Proposed for Agentic LLMs

A new arxiv paper introduces shared selective persistent memory, an architecture that identifies and retains four categories of reusable context (task specifications, data schemas, tool configurations, and tool-use patterns) for agentic LLM systems. This approach aims to solve the context problem where each session starts from zero, while avoiding token-inefficient naive persistence of entire conversation histories. The work is relevant to developers building multi-turn code generation agents.

0 engagement·1 source·arxiv
Fri, Jul 10, 2026, 03:07 PM
arXiv
Product

odek: AI agent with long-term semantic memory across sessions

odek is an AI agent that maintains a structured, semantic long-term memory across sessions, remembering user preferences, codebases, and goals without requiring re-explanation. It uses LLMs to process and store information in a three-tier memory system, solving the problem of session-only memory for users who need continuity in their interactions.

0 engagement·1 source·rss
Sat, Jul 11, 2026, 08:01 PM
RSS