📄 skillActivePromisingmedium

withastro/flue

The sandbox agent framework.

Supported Platforms

🤖 Claude Code🧠 Codex

Stars

4.9k

Skill Type

🏗️ Code Scaffolding & Templates

Quality Score

112/200

License

Apache-2.0

Forks

261

Last Updated

Jun 11, 2026

Discovered

May 10, 2026

Validation

Passed

github.com/withastro/flue

Quality Breakdown

112/ 200

Content Signals

○Gotchas/Edge Cases+40

✓Progressive Disclosure+30

○Trigger Description+20

○Verification/Safety+20

✓Code Examples+15

✓Composability+15

Repo Health

✓Recent Activity+15

✓Scripts/Automation+10

✓Real Usage (Issues)+10

✓Single Responsibility+10

✓Config/Persistence+10

✓Install Instructions+5

Multi-platform bonus: +5 pts if tool supports 2+ platforms. Score derived from 12 structural signals — not stars or popularity.

medium

Requires extended permissions (shell access, subagents). Review before use.

Active

Updated within the last 90 days. Actively maintained.

Unverified skill. Always review source code before installing any skill from an unknown author.

Framework designed for autonomous agent execution without explicit human approval gates mentioned in core architecture
Multiple sandbox connector integrations (E2B, Daytona, Modal, Vercel, Cloudflare) enabling external code execution across various platforms
Agent harness framework that spawns and orchestrates multiple agents based on AGENTS.md configuration files
File system access capabilities present in examples (fs-test, fs-surface-test agents) suggesting file modification potential
Configuration files (.flue/agents, .flue/connectors) that appear to define autonomous behaviors without visible approval workflow
Persistent configuration structure (.flue directory) that could modify execution behavior