Visual UI annotation tool for AI agents. Drop the React toolbar into any app — humans click elements and leave feedback, agents receive structured CSS selectors, bounding boxes, and React component trees to find exact code. Supports MCP watch-loop, platform-specific hooks (Claude Code / Codex / Gemini CLI / OpenCode), webhook delivery, and autonomous self-driving critique with agent-browser.
75
63%
Does it follow best practices?
Impact
96%
1.52xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agent-skills/agentation/SKILL.mdReact component integration
Dev dependency flag
0%
100%
Correct package name
100%
100%
Import from agentation
0%
100%
NODE_ENV gating
100%
100%
App Router placement
100%
100%
Endpoint prop set
0%
100%
Body-level placement
100%
100%
MCP server start command
50%
100%
Doctor verification command
0%
100%
Layout structure preserved
100%
100%
Watch loop agent instructions
Immediate acknowledge call
100%
100%
Resolve with summary
100%
100%
Blocking severity first
0%
100%
elementPath as grep target
25%
100%
reactComponents for React
100%
100%
Minimal changes only
80%
100%
Watch loop continuation
100%
100%
Watch trigger keywords
100%
100%
Dismiss requires reason
71%
100%
Watch over copy-paste
14%
42%
Multi-platform MCP configuration
Claude Code config path
0%
100%
Claude Code mcpServers key
100%
100%
Codex TOML format
0%
100%
Codex developer_instructions
40%
100%
Gemini settings.json path
100%
100%
OpenCode type local
100%
100%
OpenCode command array
62%
100%
OpenCode env vars
100%
100%
Doctor verification command
100%
100%
Universal add-mcp command
0%
0%
MCP server command consistent
100%
100%
Default port documented
100%
100%
c033769
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.