analyze-agent-behavior

Map an AI tool's system prompt to a Dojo ADA disposition YAML with confidence scores.

2|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/DojoGenesis/gateway --skill analyze-agent-behavior
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyze-agent-behavior
Source: https://github.com/DojoGenesis/gateway/tree/main/plugins/wisdom-garden/skills/analyze-agent-behavior
Command: npx skills add https://github.com/DojoGenesis/gateway --skill analyze-agent-behavior

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

External AI tools embed their behavioral models in system prompts, but these models are hidden and undocumented, making it hard to compare, replicate, or adapt their behavior within Dojo environments.

Core Features & Use Cases

  • Behavior Extraction: Scans a system prompt for pacing, depth, tone, initiative, validation, and error‑handling cues.
  • Disposition Mapping: Generates a Dojo ADA disposition YAML with confidence scores for each field.
  • Similarity Scoring: Computes a weighted similarity metric against Dojo’s default disposition.
  • Use Cases: Reverse‑engineer competitor agents, design new disposition presets, or evaluate how an external tool’s behavior aligns with Dojo standards.

Quick Start

Ask the skill to analyze a tool's system prompt and generate a disposition preset with confidence scores.

Frequently Asked Questions about analyze-agent-behavior

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

FAQPage Schema
How do I analyze an AI system prompt to map its agent behavior?

Reverse-engineering agent behavior extracts hidden behavioral models from system prompts and maps them to disposition fields, enabling you to compare, replicate, or adapt external AI tool personalities within your environment.

How do I generate a disposition YAML from an external AI tool's system prompt?

Generate a disposition YAML by applying pattern tables to the ingested system prompt text, which produces a structured configuration file containing mapped behavioral fields and calculated confidence scores.

Can I compare external AI agent behavior against default disposition standards?

Yes, you can evaluate alignment by computing a weighted similarity metric that scores an external tool's extracted behavioral patterns against default disposition standards to quantify behavioral differences.

Do I need to store the system prompt in MemoryStore before analyzing agent behavior?

Yes, analyzing agent behavior requires accessing the ingested system prompt stored in MemoryStore, as the pattern extraction and disposition mapping logic depends on this stored text to function.

What are the limitations of reverse-engineering tool personalities from system prompts?

Limitations include reliance on explicit textual cues within the prompt, meaning undocumented implicit behaviors or dynamically adjusted runtime parameters may not be captured accurately in the generated disposition YAML.