Tilde.run: An Agent Sandbox with a Transactional, Versioned Filesystem

Tilde.run is a sandbox for AI agents that turns every agent run into a reversible transaction. It creates a single versioned filesystem from multiple data sources — GitHub, S3, and Google Drive are mounted as a single ~/sandbox directory. Every file is versioned from the first commit, and any agent run can be rolled back instantly with one command.
Key Features
Versioned Composable Filesystem
A real POSIX filesystem—any tool, any language, no SDKs. Mount code from GitHub, training data from S3, and documents from Google Drive as a single ~/sandbox. Example mounts:
4 mounts
~/sandbox
├── code github acme/ml-pipeline
├── data s3 acme-data/training
├── docs gdrive team-wiki
└── output local
All versioned, all reversible.
Safe Serverless Sandboxes
Each run is a transaction in a fresh, isolated container. On clean exit, changes commit atomically; on failure, nothing changes. No backups to restore, no manual cleanup.
Network Isolation
Cloud metadata, private networks, and unauthorized hosts are blocked by default. Every outbound request is policy-checked and logged. Example egress log:
12:04:01 GET api.openai.com /v1/completions ALLOW
12:04:03 POST api.anthropic.com /v1/messages ALLOW
12:04:05 GET pypi.org /simple/pandas ALLOW
12:04:07 POST evil-exfil.io /upload DENY
12:04:08 GET 169.254.169.254 /metadata DENY
12:04:09 PUT registry.npmjs.org /my-pkg DENY
Time Travel & Audit Trail
Browse the full timeline, inspect diffs, and revert any commit instantly. Every change is tied to the human, process, or agent that produced it.
Agent-first RBAC
Agents have scoped permissions—never full user access. Granular policies in a simple DSL. Example policy:
analyst-policy
GetObject (path: "/data/*") ?
PutObject (path: "/reports/*") # require human approval!
! PutObject (path: "/secrets/*")
Quickstart
Install in one line:
$ curl -fsSL https://tilde.run/install | sh
Run an agent in a sandbox:
$ tilde exec my-team/documents \
--image python:3.12 \
-- /sandbox/code/agent.py --input /sandbox/data/reports
sandbox running...
sandbox completed. exit code: 0, commit id: c9d0e1f2
Or start an interactive shell:
$ tilde shell my-team/documents --image python:3.12
root@sb-7f3a9c01:/sandbox$ _
Python SDK:
import tilde
repo = tilde.repository("my-team/documents")
with repo.shell(image="python:3.12") as sh:
sh.run("pip install pandas")
result = sh.run("python agent.py --input /sandbox/data")
print(result.stdout.text())
Integrates with Claude, AWS S3, LangGraph, Google Drive, and Hugging Face.
📖 Read the full source: HN AI Agents
👀 See Also

Agent Image Skills: Simple Image Hosting for Claude Code Agents
A developer built a simple image hosting service at https://images.labnocturne.com to solve the problem of Claude Code agents generating images with nowhere to store them. The service provides instant test keys via curl, auto-deletes files after 7 days, and includes Claude Code MCP skills for upload, file listing, and deletion.

Flavian: A WordPress Development Framework with 24 Specialized Claude Code Agents
Flavian is an open-source WordPress development framework built around Claude Code, featuring 24 specialized agents for tasks like frontend development, security audits, and Figma-to-WordPress conversion. The creator found domain-specific agents significantly outperform general-purpose ones for WordPress development.

Open Source Curated Collection of OpenClaw Resources Unveiled
Discover a new open-source collection of OpenClaw resources, curated by the community to enhance AI development and collaboration.

Clawback: Hooks-based implementation of leaked Claude verification loops
Clawback is a GitHub project that reimplements the verification loops from the Claude source map leak as mechanical hooks instead of prompts. It includes stop hooks, PreToolUse, PostToolUse, and PostCompact hooks that can't be skipped by the model under context pressure.