langshell

What problem does it solve?

This Skill eliminates unsafe and over-permissioned tool use by letting AI agents run multi-step Python or TypeScript logic inside a persistent sandbox where all side effects are gated by explicitly registered capabilities.

Core Features & Use Cases

• Stateful session execution: Keep variables, caches, and helper functions across turns via persistent sessions.
• Capability-scoped external access: Read/write files, fetch HTTP, and query databases only through host-registered functions (e.g., read_text, write_text, fetch_json).
• Pre-run validation and safe error handling: Use validate/dry-run to catch unsupported features, unknown tools, and permission issues before side effects.
• Structured results: Prefer returning JSON-serializable objects to produce reliable RunResult output for downstream systems.

Use case example: In an agent workflow that must transform data end-to-end, use LangShell to fetch approved JSON, filter and aggregate it, write a sanitized report into an authorized mount, and return a compact summary object.

Quick Start

Use the langshell skill to compute a safe, structured result and call only approved capabilities by writing Python or TypeScript code that assigns the final value to result.