AgentIndex icon
AgentIndex
sema logo
sema
★ 54
vs
leather logo
leather
★ 17

sema vs leather

sema: Sema is a Lisp-like language designed to abstract away the common scaffolding of LLM agent loops, making agents and LLM calls language primitives rather than relying on SDKs. It compiles to a fast bytecode VM and ships as a single binary, enabling efficient development and deployment of LLM-powered applications.; leather: Leather provides a local agent infrastructure in a single Go binary, enabling declarative agents to run scheduled jobs, webhook-driven workflows, and tool calling without a Python stack or external dependencies. It includes features like multi-stage agent routing, queues, and developer tools for comprehensive LLM-powered workflow automation.

01

TL;DR

sema logo
semaWorkflow Automation · ★ 54

Building coding agents capable of file manipulation, code review, and shell command execution.

Read sema full review →
leather logo
leatherWorkflow Automation · ★ 17

Scheduled Automation: Run agents on a predefined schedule for routine tasks.

Read leather full review →
02

Side-by-Side Comparison

Field
sema logosema
leather logoleather
Category
Workflow Automation
Workflow Automation
Stars
★ 54
★ 17
License
MIT
GPL-3.0
Install Command
curl -fsSL https://sema-lang.com/install.sh | sh
go install github.com/TGPSKI/leather/cmd/leather@latest && go install github.com/TGPSKI/leather/cmd/shell-mcp@latest
Updated
1d ago
1mo ago
Open Source
Yes
Yes
Website
↗ Visit
↗ Visit
GitHub
↗ GitHub
↗ GitHub
Tags
Lisp, LLM Agents, Workflow Automation
Go, LLM Agents, Workflow Automation
03

Features

sema logosema
01Treats LLMs as language primitives, with prompts, messages, conversations, tools, and agents as first-class data types.
02Supports multiple LLM providers (e.g., Anthropic, OpenAI, Gemini) with automatic configuration via environment variables.
03Includes built-in pipeline features such as response caching, fallback chains, rate limiting, retry mechanisms, and text chunking.
04Provides cost-aware development with integrated budget tracking and bundled pricing snapshots for LLM interactions.
05Offers a comprehensive developer toolchain, including a formatter, language server (LSP), debugger (DAP), and an MCP server for LLM clients.
leather logoleather
01Agent Definition: Define agents using Markdown with system prompts and variable support.
02Tool Calling: Integrate shell commands and other skills as agent tools.
03Multi-Agent Curing: Bind agents to queues for flexible multi-stage routing and complex workflows.
04Persistent Queues: Utilize filesystem FIFO queues with configurable depth, backpressure, and concurrency.
05Developer UI (DevTools): A web-based UI with session timeline, event inspector, and live updates for monitoring workflows.
04

Use Cases

sema logosema
↳Building coding agents capable of file manipulation, code review, and shell command execution.
↳Automating tasks like generating conventional commit messages from staged Git changes.
↳Performing structured data extraction from text and images, and conducting multi-modal chat interactions.
↳Implementing Retrieval-Augmented Generation (RAG) pipelines using in-memory vector stores and cross-encoder reranking.
↳Developing interactive LLM-powered notebooks for rapid prototyping and experimentation.
leather logoleather
↳Scheduled Automation: Run agents on a predefined schedule for routine tasks.
↳Event-Driven Workflows: Trigger agent runs via webhooks for reactive automation.
↳Multi-Agent Pipelines: Orchestrate multiple agents into complex, multi-stage processing pipelines.
↳Local LLM Integration: Integrate local LLM endpoints (e.g., Ollama) for privacy-preserving AI tasks.
↳DevTools for Monitoring: Visualize and debug agent execution and workflow events in real-time.
FAQ

FAQ

What is the difference between sema and leather?
Both sema and leather are in the Workflow Automation category. sema has 54 stars, while leather has 17 stars.
Which is better, sema or leather?
The best choice depends on your use case. Choose sema if Building coding agents capable of file manipulation, code review, and shell command execution., and leather if Scheduled Automation: Run agents on a predefined schedule for routine tasks..
Is sema free or open source?
Yes, sema is open source on GitHub (MIT).
Is leather free or open source?
Yes, leather is open source on GitHub (GPL-3.0).
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
→

Related

Alternatives to sema →Alternatives to leather →sema details →leather details →