Skip to content

Agent Coach

United States flag

Agent Coach — Security-scanned skills and a daily coach for AI agents

DR1: compare Agent Coach's Domain Rating with other productsAICompare products for Agent Coach
Visit site

About Agent Coach

Agent Coach is a free training coach for AI agents. An owner pastes one line into their agent. The agent describes its goal and gets matched to skills, GitHub repos, MCP servers and prompts from a library where every item passed a security scan for prompt injection, malicious code and secret exfiltration. The agent checks in daily. Ten public verdict rules decide whether each item becomes permanent, stays temporary for one task, or is marked unproven and pending removal. It works over MCP or a single fetchable instruction URL. Made by Avatar 8 LLC.

Who it is for
Owners of AI agents such as Claude Code, Cursor, Codex, OpenClaw or Grok Bot
Problem it solves
Agents need skills, GitHub repos, MCP servers and prompts to reach their goals, and each item should be checked for prompt injection, malicious code and secret exfiltration before the agent uses it.
The solution
Paste one line into your agent. It declares one goal, gets matched with scanned skills, MCP servers, repos and prompts, and checks in daily. The coach says what to keep and what to drop.
What makes it unique
Every item in the library passed a security scan. Ten public verdict rules decide whether each item becomes permanent, stays temporary for one task, or is marked unproven and pending removal.

Features

  • Security-scanned library

    Skills, repos, MCP servers and prompts passed a security scan.

Details

Platforms
  • API
  • Web
Pricing
  • Free
Built for

Owners of AI agents such as Claude Code, Cursor, Codex, OpenClaw or Grok Bot

Building since

2026

Technology

Agent Coach product screenshot

1 / 1

Website performance

Leaderboard

Recorded mobile lab measurements and finalized weekly standings from IndieTools Speed.

92ExcellentMobile Lighthouse

Measured · UTC

Load time
690 ms
FCP
1.20 s
LCP
3.13 s
CLS
0.000
TTFB
51 ms
TBT
21 ms
Speed index
3.36 s

These stored lab observations describe controlled test runs. They are not real-user Core Web Vitals or a guarantee of search rankings. Weekly standings below remain attached to their finalized edition.

Overall rank
Not finalized
Best overall rank
Not finalized
0 top 10 finishes
Best category rank
Not finalized
0 category wins
Latest change
-4 score
No finalized score yet

Method: psi-mobile-two-sample-mean-v1 · 2 samples per reading. Only readings with the same recorded method and sample count share a chart. Lab results can vary between runs.

Website performance history

Inspect every successful mobile Lighthouse lab measurement by metric and date range.

Range

Latest score

92 / 100

Higher is better
5 measurements

Use the left and right arrow keys to inspect measurements. Use Home or End to jump to the first or last measurement. Press Escape to dismiss the chart tooltip.

Oct 25, 2026Oct 27, 2026

Related products

Browse AI

Others in the same categories.

Visit site