scottgl9

scottgl9 / skelm

Public

skelm - open-source framework for secure, agentic, long-running workflows. Authoring, scheduling, and operating typed pipelines that mix deterministic code, LLM inference, and agent loops under default-deny permissions.

scottgl9.github.ioskelm acp agent agentic agents automation
13
6
100% credibility
Found May 06, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Skelm is a TypeScript framework for creating secure, long-running workflows that combine code, AI calls, and agent interactions, runnable on demand, scheduled, or as persistent services.

How It Works

1
🔍 Discover skelm

You hear about a friendly tool for building safe automations that mix everyday tasks with smart AI helpers.

2
📥 Grab the helper tool

Download the simple command-line buddy that makes everything easy.

3
✨ Create your first automation

Set up a ready example project that greets people by name.

4
▶️ Test your creation

Run it once with your own details and see the magic happen.

5
Choose how to use it
⚡
Quick one-time run

Fire it off whenever you need a fast result.

⏰
Set repeating schedule

Make it happen automatically on a timer.

🏠
Launch always-ready service

Start a background helper that waits for your calls.

🎉 Your automations thrive

Enjoy secure, reliable helpers handling tasks like chat bots or daily summaries without worry.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 13 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is skelm?

Skelm is a TypeScript framework for authoring secure, agentic, long-running workflows that mix deterministic code, LLM inference, and agent loops under default-deny permissions. It solves the pain of hand-rolling scripts for tasks like GitHub triage or Slack digests by letting you define typed pipelines, schedule them with cron or webhooks, and run them via CLI or a persistent gateway service. Developers get refactorable TypeScript modules, persistent workspaces per agent, and tamper-evident audit logs, all local-first with SQLite or Postgres.

Why is it gaining traction?

Unlike LangChain or CrewAI, skelm enforces default-deny security structurally—no plugins needed—and skips DSLs for plain TypeScript you can test and version like any code. Multi-backend support for Anthropic, OpenAI, and MCP-native agents stands out, plus no telemetry and a gateway with HTTP/SSE endpoints for production. The hook is building robust, auditable agentic loops for Skelmersdale-style ops without cloud lock-in.

Who should use this?

Backend engineers building agentic workflows for GitHub PRs, Jira queues, or email triage under tight permissions. DevOps teams scheduling LLM-enriched nightly reports to Slack or Discord. Anyone tired of fragile cron + LLM scripts needing secure, long-running determinism like Skjelmose cyclist's training loops.

Verdict

Promising for secure TypeScript agent workflows, with solid docs, recipes, and CLI (skelm init/run/gateway), but at 13 stars and 1.0% credibility, it's early—APIs unstable until v1. Try for prototypes if security matters; star and contribute to shape it.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.