ai-training-mode

Collect client feedback and inject training data into agent prompts.

Updated Feb 24, 2026
One-click install
npx skills add https://github.com/teodorboev/socialai --skill ai-training-mode
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-training-mode
Source: https://github.com/teodorboev/socialai/tree/main/.opencode/skills/ai-training-mode
Command: npx skills add https://github.com/teodorboev/socialai --skill ai-training-mode

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill allows clients to actively train AI agents by providing feedback on their outputs, leading to increasingly personalized and accurate AI performance over time, thereby increasing client retention.

Core Features & Use Cases

  • Feedback Collection: Gathers various forms of user feedback including ratings, corrections, and explicit preferences.
  • Prompt Injection: Integrates learned preferences and corrections directly into agent prompts for real-time adaptation.
  • Pattern Learning: Automatically identifies recurring patterns from feedback to derive new preferences.
  • Use Case: A client consistently rates LinkedIn posts with emojis poorly. The AI learns this preference and stops adding emojis to their LinkedIn content, making the AI's output more aligned with the client's brand.

Quick Start

Use the ai-training-mode skill to provide feedback on the AI's last output.

Frequently Asked Questions about ai-training-mode

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fine-tune AI agents using client feedback?

Personalize AI outputs by collecting direct client feedback such as ratings, corrections, and explicit rules, then injecting this per-organization data into agent prompts to improve output quality and personalization over time.

What is prompt injection for AI personalization?

Prompt injection for AI personalization integrates learned client preferences and corrections directly into agent prompts, enabling real-time adaptation and increasingly aligned outputs based on stored feedback data.

How do I collect user feedback to improve prompt engineering?

Collect user feedback to improve prompt engineering by gathering ratings, corrections, and explicit preferences, then automatically identifying recurring patterns to derive new preferences for agent enhancement.

Does this feedback loop support per-organization training data?

The feedback loop supports per-organization training data by storing client feedback, preferences, and exemplars via a structured database schema, ensuring personalized agent enhancement for each specific organization.

Can I use direct feedback mechanisms to stop unwanted AI behaviors?

Use direct feedback mechanisms like consistent low ratings to stop unwanted AI behaviors, as the system learns from recurring patterns and automatically derives preferences to adjust future agent outputs accordingly.