Skip to main content

SLOP

Structured Language for Orchestrating Prompts

A sandboxed execution environment for LLM-generated code. Scripts pause mid-run; the code, call stack, and memory are saved as editable JSON โ€” fix anything, then resume with no work lost.

# Chain MCP calls, pause, edit, resume
repos = github.search(query: "mcp servers")
pause("after_fetch")   # state saved as editable JSON
result = llm.call(prompt: "Summarize: " + json_stringify(repos), schema: {summary: string})
emit(result.summary)

๐ŸŽฏ Simple & Intuitive

Python-like syntax that's easy to learn. Write AI workflows in minutes, not hours. No complex boilerplate or configuration required.

๐Ÿ”’ Safe by Design

Built-in safety mechanisms prevent infinite loops, resource exhaustion, and unintended side effects. Perfect for production AI systems.

โšก Fast & Efficient

Lightweight Go runtime with streaming support. Execute complex agent workflows with minimal overhead and real-time responses.

๐Ÿ”Œ MCP Integration

Native Model Context Protocol support for tool calling and external service integration. Build powerful AI agents with ease.

๐Ÿ“ฆ Modular Design

Organize code into reusable modules. Share agents across projects with built-in dependency management and versioning.

๐Ÿงช Well-Tested

200+ unit tests ensure reliability. Battle-tested patterns for common AI workflows and edge cases.