S
pending
"Awesome LLM Apps: 100+ Free, Open-Source AI Agents You Can Clone and Ship Today"
Grounded / Real
Inflated / Uruttu
Original Content
100+ open-source AI agents and agent skills you can clone and ship today 🤯
Awesome LLM Apps with 123,000+ stars is trending again on GitHub.
It's 100% free and Open Source. So you can clone it, ship it, even sell it.
Others charge $20,000+ for courses that teach less than what's sitting in this one repo.
Every agent is fully-built and tested end to end. Pick one, add your API key, and it runs in 30 seconds.
Some of the newest AI Agents and Agent skills inside:
1. Project Graveyard: an agent skill that autopsies your dead side projects and tells you why each one died
2. Self-Improving Agent Skills: skills that rewrite themselves against evals using Gemini and ADK
3. Insurance Claim Live Agent Team: voice claims settled in real time with Live Agent teams
4. Advisor Orchestrator Worker: a meta loop where one model advises, one orchestrates, one executes
5. Always-on HN Briefing Agent: reads Hacker News while you sleep, ships you a ranked brief
6. AI Fraud Investigation Agent: cross-examines public records to flag facilities that don't add up
7. AI Home Renovation Agent: photo of your space in, photorealistic redesign out
The agent skills install into Claude Code, Codex, or Cursor with one command. Plain English to use.
Works with Claude, Gemini, GPT, DeepSeek, Llama, and Qwen. Run local or in the cloud.
New agents and skills drop every week. Star it now, clone one this weekend.
Link to the GitHub repo in the comments.
____
100+ AI Agents, Agent Skills, and Agent teams (free and open-source with 125,000+ stars): https://lnkd.in/dW6b_dEn
♻️ Repost to share it with your network. Follow me at Shubham Saboo for more such AI Agent, Loop Engineering and Agent Skill workflows.
Awesome LLM Apps with 123,000+ stars is trending again on GitHub.
It's 100% free and Open Source. So you can clone it, ship it, even sell it.
Others charge $20,000+ for courses that teach less than what's sitting in this one repo.
Every agent is fully-built and tested end to end. Pick one, add your API key, and it runs in 30 seconds.
Some of the newest AI Agents and Agent skills inside:
1. Project Graveyard: an agent skill that autopsies your dead side projects and tells you why each one died
2. Self-Improving Agent Skills: skills that rewrite themselves against evals using Gemini and ADK
3. Insurance Claim Live Agent Team: voice claims settled in real time with Live Agent teams
4. Advisor Orchestrator Worker: a meta loop where one model advises, one orchestrates, one executes
5. Always-on HN Briefing Agent: reads Hacker News while you sleep, ships you a ranked brief
6. AI Fraud Investigation Agent: cross-examines public records to flag facilities that don't add up
7. AI Home Renovation Agent: photo of your space in, photorealistic redesign out
The agent skills install into Claude Code, Codex, or Cursor with one command. Plain English to use.
Works with Claude, Gemini, GPT, DeepSeek, Llama, and Qwen. Run local or in the cloud.
New agents and skills drop every week. Star it now, clone one this weekend.
Link to the GitHub repo in the comments.
____
100+ AI Agents, Agent Skills, and Agent teams (free and open-source with 125,000+ stars): https://lnkd.in/dW6b_dEn
♻️ Repost to share it with your network. Follow me at Shubham Saboo for more such AI Agent, Loop Engineering and Agent Skill workflows.
Validated Content
All seven listed agents/skills are confirmed real entries in the repository. This is a very accurate post.
✅ Confirmed — every specific claim checks out against the actual repo
"Awesome LLM Apps" repo by Shubham Saboo is real — confirmed, GitHub: Shubhamsaboo/awesome-llm-apps, "100+ AI Agents, Agent Skills and RAG Apps."
123,000+ / 125,000+ stars — plausible and consistent with growth trajectory. The author's own X profile bio (checked around the same period) states "127,000+ stars," so 123k-125k fits neatly as a slightly earlier snapshot in a fast-growing, actively-trending repo.
100% free and open source, clone/ship/sell it — confirmed. The repo is Apache-2.0 licensed, and its own README explicitly says "Clone it, ship it, sell it."
All seven named agents/skills are real, exact matches:
Project Graveyard — confirmed exactly ("Finds every side project you abandoned, tells you why each one died...")
Self-Improving Agent Skills — confirmed ("Automatically optimize agent skills using Gemini and ADK")
Insurance Claim Live Agent Team — confirmed ("Real-time voice claim intake with Gemini Live")
Advisor Orchestrator Worker — confirmed almost verbatim ("Meta Loop with Claude Fable 5 as advisor, GPT-5.6 as orchestrator, and Gemini 3.5 Flash as worker")
Always-on HN Briefing Agent — listed in the repo's gallery
AI Fraud Investigation Agent — confirmed ("public records, cross-examined")
AI Home Renovation Agent — confirmed ("photo in, photoreal redesign out")
Agent skills install via one command into Claude Code, Codex, Cursor — confirmed: the repo documents npx skills add [url] as the install method, and states skills work with "Claude Code, Codex, Cursor, and other coding agents."
Works with Claude, Gemini, GPT, DeepSeek, Llama, Qwen — confirmed, matches the repo's stated model compatibility exactly.
New agents/skills drop weekly — confirmed, matches the repo's own "New templates drop weekly" messaging.
⚠️ Minor stylistic note, not a factual error
"Others charge $20,000+ for courses that teach less than what's sitting in this one repo" — this is a subjective marketing comparison, not a verifiable fact. Reasonable as opinion/hype, but not something to fact-check.
Bottom line: This is an unusually accurate promotional post — every named agent/skill, the install mechanism, the model compatibility, and the open-source/licensing claims all match the actual repository precisely. The only non-factual element is the comparative "$20,000 courses" hype line, which is subjective framing rather than a checkable claim.
✅ Confirmed — every specific claim checks out against the actual repo
"Awesome LLM Apps" repo by Shubham Saboo is real — confirmed, GitHub: Shubhamsaboo/awesome-llm-apps, "100+ AI Agents, Agent Skills and RAG Apps."
123,000+ / 125,000+ stars — plausible and consistent with growth trajectory. The author's own X profile bio (checked around the same period) states "127,000+ stars," so 123k-125k fits neatly as a slightly earlier snapshot in a fast-growing, actively-trending repo.
100% free and open source, clone/ship/sell it — confirmed. The repo is Apache-2.0 licensed, and its own README explicitly says "Clone it, ship it, sell it."
All seven named agents/skills are real, exact matches:
Project Graveyard — confirmed exactly ("Finds every side project you abandoned, tells you why each one died...")
Self-Improving Agent Skills — confirmed ("Automatically optimize agent skills using Gemini and ADK")
Insurance Claim Live Agent Team — confirmed ("Real-time voice claim intake with Gemini Live")
Advisor Orchestrator Worker — confirmed almost verbatim ("Meta Loop with Claude Fable 5 as advisor, GPT-5.6 as orchestrator, and Gemini 3.5 Flash as worker")
Always-on HN Briefing Agent — listed in the repo's gallery
AI Fraud Investigation Agent — confirmed ("public records, cross-examined")
AI Home Renovation Agent — confirmed ("photo in, photoreal redesign out")
Agent skills install via one command into Claude Code, Codex, Cursor — confirmed: the repo documents npx skills add [url] as the install method, and states skills work with "Claude Code, Codex, Cursor, and other coding agents."
Works with Claude, Gemini, GPT, DeepSeek, Llama, Qwen — confirmed, matches the repo's stated model compatibility exactly.
New agents/skills drop weekly — confirmed, matches the repo's own "New templates drop weekly" messaging.
⚠️ Minor stylistic note, not a factual error
"Others charge $20,000+ for courses that teach less than what's sitting in this one repo" — this is a subjective marketing comparison, not a verifiable fact. Reasonable as opinion/hype, but not something to fact-check.
Bottom line: This is an unusually accurate promotional post — every named agent/skill, the install mechanism, the model compatibility, and the open-source/licensing claims all match the actual repository precisely. The only non-factual element is the comparative "$20,000 courses" hype line, which is subjective framing rather than a checkable claim.