# The Missing Abstraction for AI Agents: The Agent Filesystem

DevFeed: [The Missing Abstraction for AI Agents: The Agent Filesystem](<https://devfeed.tech/articles/the-missing-abstraction-for-ai-agents-the-agent-filesystem-5881.md>)

Original publisher: [Read original article](<https://turso.tech/blog/agentfs>)

Author: Pekka Enberg

Published: 2025-11-13T00:00:00Z

Content type: article

Language: en

Sources: [Turso Blog](<https://devfeed.tech/sources/turso-blog.md>)

Topics: [AI Bots](<https://devfeed.tech/topics/ai-bots.md>), [ai observability](<https://devfeed.tech/topics/ai-observability.md>), [AI, ML & Data Engineering](<https://devfeed.tech/topics/ai-ml-data-engineering.md>), [SDKs](<https://devfeed.tech/topics/sdks.md>), [debugging](<https://devfeed.tech/topics/debugging.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [observability](<https://devfeed.tech/tags/observability.md>), [sandboxes](<https://devfeed.tech/tags/sandboxes.md>), [sql](<https://devfeed.tech/tags/sql.md>), [sqlite](<https://devfeed.tech/tags/sqlite.md>)

## AI overview

AgentFS is an SDK that gives AI agents a unified SQLite-backed filesystem for artifacts, key-value state, and tool-call tracking. It is designed to make agent state portable, queryable, auditable, snapshotable, and reproducible.

## Source excerpt

AgentFS gives AI agents a single SQLite-backed filesystem for files, key-value state, and tool call audit trails. One file you can snapshot and query.