# agentleFS > agentleFS is a permissioned, versioned store of files that people and their AI agents use together over MCP. Each agent reads only what the person behind it can read. Shared second brain for any AI with agent permissions. Teammates choose their models, providers and harnesses while sharing context, work and the reasoning behind decisions. Each agent works within the access it receives. agentleFS is a permissioned file store that AI agents read over MCP (Model Context Protocol). Access is decided per PERSON rather than per API key: an agent connected on someone's behalf reads exactly what that person is allowed to read, and the same authorization filter runs in the SQL WHERE clause before ranking, so content the caller cannot reach is never returned, listed or matched. ## Pages - [Team and AI agent collaboration — agentleFS](https://agentlefs.com/): Shared context, decisions and work across models, providers and harnesses. Let teammates choose their AI tools and collaborate from one permissioned workspace. - [Collaboration across agents and AI tools — agentleFS](https://agentlefs.com/solutions): Coordinate people and agents with shared documents, claims, handoffs and durable waits. Explore shared knowledge, decision history, skills and access controls. - [Coordinate your team’s agents — agentleFS](https://agentlefs.com/solutions/collaboration): Compare approaches in a team ballot, prepare a work brief, and coordinate agents with claims, handoffs, durable waits and progress reports. - [Shared knowledge for every AI — agentleFS](https://agentlefs.com/solutions/knowledge): Bring project notes, linked documents, GitHub and Drive sources, and shared skills into one place your team’s agents can work from. - [Agent access and sharing — agentleFS](https://agentlefs.com/solutions/access): Choose roles on folders and files, narrow each agent’s delegated scope, and review access without handing over your whole workspace. - [CLAUDE.md vs AGENTS.md — agentleFS](https://agentlefs.com/context/agents-md-vs-claude-md): Compare CLAUDE.md and AGENTS.md with counts collected from public GitHub projects. Browse examples of both in the agentleFS instruction corpus. - [Blog: Agent memory and agent context | agentleFS](https://agentlefs.com/blog): Engineering persistent agent memory and useful agent context: retrieval, lifecycle hooks, shared knowledge, and lessons from real instruction files. - [How-to guides for agent memory and context | agentleFS](https://agentlefs.com/blog/how-to): Step-by-step guides to agent memory and context: writing CLAUDE.md and AGENTS.md files, checking them, and the rules worth enforcing with hooks. - [Meta harness: async agent collaboration | agentleFS](https://agentlefs.com/blog/meta-harness-async-agent-collaboration): Our coming meta harness connects agent sessions with a shared UX, async callbacks and permissioned cross-harness work. Read the approach and join the waitlist. - [Private AI: keep your competitive advantage | agentleFS](https://agentlefs.com/blog/private-ai-local-llm): Is ChatGPT safe for confidential information? Private AI needs private data and private inference. Here's how to get both, with a local LLM. - [Persistent AI Agent Memory: MCP and Hooks | agentleFS](https://agentlefs.com/blog/persistent-ai-agent-memory-lifecycle-hooks): How persistent AI agent memory moves from optional MCP calls to lifecycle hooks, with retrieval deadlines, durable capture, and honest failure policies. - [External Agent Memory Across Clients | agentleFS](https://agentlefs.com/blog/agent-memory-outside-the-harness): Why external agent memory enables cross-client continuity, deliberate sharing across tenants, and a retrieval stack you can tune and evaluate. - [CLAUDE.md best practices from 5,786 real files — agentleFS](https://agentlefs.com/blog/claude-md-best-practices): CLAUDE.md and AGENTS.md best practices from 5,786 real files: what people cut as models improved, what they kept, and what to enforce with hooks. - [Agent swarm communication with shared docs | agentleFS](https://agentlefs.com/blog/agent-swarm-communication): Doorbell-and-cabinet swarms on agentleFS: coordinator–worker, shared docs, cron heads. How the wake works, what we measured, and a Product→Admin loop. - [ChatGPT to Claude, and back, without losing work | agentleFS](https://agentlefs.com/blog/chatgpt-to-claude): Switch between ChatGPT, Claude and Gemini without copying and pasting: connect every AI to one agentleFS folder, and each one reads the same plans. - [Claude Code agent teams that share files safely | agentleFS](https://agentlefs.com/blog/claude-code-agent-teams): Claude Code subagents inherit your MCP tools and permission mode. Give each one its own agentleFS identity, scoped to one folder and ended with one call. - [Claude Code memory: your personal CLAUDE.md | agentleFS](https://agentlefs.com/blog/claude-code-memory): Your personal CLAUDE.md lives on one laptop and your team's lives in the repo. Keep yours where every AI reads it, on every machine, outside the repo. - [CLAUDE.md and AGENTS.md examples from real repos | agentleFS](https://agentlefs.com/blog/claude-md-examples): Twelve real CLAUDE.md and AGENTS.md examples from well-starred repos, the lines worth copying, and a short template. The good ones are short. - [Claude skills in every AI, shared with your team | agentleFS](https://agentlefs.com/blog/claude-skills): Claude skills live on one disk or in one repo. Keep them in one agentleFS folder that Claude, Codex, Gemini and Grok all read, and share it with your team. - [Obsidian + Claude: a second brain every AI reads | agentleFS](https://agentlefs.com/blog/obsidian-claude-second-brain): Obsidian keeps your vault on your own devices, where no AI can read Sync's copies. Put it in agentleFS, and Claude, ChatGPT and Gemini read it anywhere. - [Share a Claude project with anyone, and their AI | agentleFS](https://agentlefs.com/blog/share-claude-project): Claude shares a project only inside one Team or Enterprise org. Here's how to share one with anyone, and let each person's own AI read and write it. - [Documentation — agentleFS](https://agentlefs.com/docs): Connect an MCP client to agentleFS, then read how permissions, versions and organizations decide what an agent can reach. - [Questions about agent access and sharing — agentleFS](https://agentlefs.com/docs/faq): Answers people look for before and after connecting an agent: how search works, who can read a folder, and how sharing works inside your organization. - [Folders and version history — agentleFS](https://agentlefs.com/docs/folders): How agentleFS stores content in folders and files, keeps every document's own version history, and lets permissions inherit down the folder tree. - [People, agents, tokens and grants — agentleFS](https://agentlefs.com/docs/identity): A token authenticates a person or an agent and carries no access of its own. What it can read comes from grants that are checked on every request. - [Team ballots, handoffs and shared work — agentleFS](https://agentlefs.com/docs/collaboration): Choose an approach with a team ballot, then coordinate agents through shared documents, claims, handoffs, durable waits and execution reports. - [Connect your harness: plugin or MCP — agentleFS](https://agentlefs.com/docs/connect): Add agentleFS to Claude Code, Claude Desktop, Cursor or any MCP client. You sign in through the browser, and the agent reads what you are allowed to read. - [REST API for documents and search — agentleFS](https://agentlefs.com/docs/api): Call agentleFS from your own code over HTTP. Read, write, move and search documents with an API key, under the same permissions an agent receives. - [Console reference: files, access, audit — agentleFS](https://agentlefs.com/docs/reference): What each part of the agentleFS console does: browsing and editing files, granting access to people and agents, mirroring sources, and the audit log. - [Privacy policy — agentleFS](https://agentlefs.com/privacy): What agentleFS stores, how long it is kept, and which third parties process it. Includes how Google user data from connected accounts is handled. - [Support — agentleFS](https://agentlefs.com/support): How to get help with agentleFS: the address that reaches us, the docs that answer the most common questions, and where to file a plugin issue. - [Terms of service — agentleFS](https://agentlefs.com/terms): The terms covering an agentleFS account, the content you upload to it, and the agents you connect to it on your organization's behalf. ## For agents - [Connect an agent to agentleFS](https://agentlefs.com/connect.md): Install instructions written for the agent itself. It runs the install for Claude Code, Codex or Cursor, then hands the browser sign-in to the person. ## Contact Write to contact@agentlefs.com. A person reads it. ## Questions and answers The full FAQ is inline at [llms-full.txt](https://agentlefs.com/llms-full.txt), 57 entries covering connecting a client, the permission model, sharing inside your organization, and what happens to your content.