rules-capture

Detect implicit user behavioral corrections and preferences in AI interactions.

5|Updated Feb 16, 2026
One-click install
npx skills add https://github.com/technomensch/knowledge-graph --skill rules-capture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rules-capture
Source: https://github.com/technomensch/knowledge-graph/tree/main/skills/rules-capture
Command: npx skills add https://github.com/technomensch/knowledge-graph --skill rules-capture

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill identifies implicit behavioral corrections and preference statements in user prompts and offers to record them as authoritative rules, streamlining behavioral consistency and process standardization across sessions.

Core Features & Use Cases

  • Behavioral Rule Detection: Recognizes phrases like "always", "never", "from now on" to identify session-wide directives.
  • Preference Capture: Captures user preferences such as "I prefer" or "I like" for personalized behavior adjustments.
  • Use Case: A user notes, "Always review the document before sending" during a session; this Skill detects such prompts and suggests recording the rule for future adherence, saving manual tracking.
  • Platform Context Detection: Analyzes whether the rule applies in Claude Code, Gemini, Cursor, Windsurf, or is platform-agnostic.

Quick Start

Use natural language instructions like "Make that a rule" to initiate rule recording, with automatic detection and routing to the appropriate configuration files based on detected platform and scope.

Frequently Asked Questions about rules-capture

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

FAQPage Schema
How do I save behavioral rules and preferences automatically in AI workflows?

To save behavioral rules automatically in AI workflows, you can use natural language phrases like "always" or "I prefer" during your session. The system detects these implicit corrections and offers to record them as authoritative rules for standardized behavior.

Does this behavioral rule capture work across different platforms like Claude and Gemini?

Yes, behavioral rule capture works across various platform environments including Claude, Gemini, Cursor, and Windsurf. It detects the platform context to determine whether a rule applies specifically to one platform or is platform-agnostic.

What is the best way to ensure consistent AI behavior across multiple sessions?

The best way to ensure consistent AI behavior across sessions is by capturing implicit behavioral corrections as authoritative rules. This process standardizes behavior management by validating instructions against safety and relevance criteria before recording them.

How do I make a rule from a preference I expressed during an AI chat session?

To make a rule from a chat preference, simply use natural language instructions like "Make that a rule" or state your preferences directly. The system automatically detects these statements and routes them to the appropriate configuration files.

Are there limitations on what types of instructions can be recorded as session rules?

Yes, limitations exist regarding what instructions can be recorded as session rules. The system validates all detected behavioral corrections and preference statements against safety and relevance criteria before recording them to ensure compliance.