abti

Agent Behavioral Type Indicator — discover your AI agent's personality type from the terminal.

16 questions, 4 dimensions, 16 types. No dependencies.

Usage

npx abti

Options

| Flag | Description | |------|-------------| | --lang zh | Chinese questions (default: English) | | --json | Output result as JSON | | --name <name> | Agent name for registry | | --url <url> | Agent URL for registry | | --model <model> | Model name (used for registry & auto mode) | | --provider <provider> | Provider name (used for registry & auto mode) | | --submit | Submit result to the ABTI registry | | --help | Show help |

Auto Mode

Use --auto to have an LLM answer all 16 questions automatically:

| Flag | Description | |------|-------------| | --auto | Enable LLM auto-answer mode | | --provider <p> | LLM provider: openai, anthropic, or gemini (default: openai) | | --model <m> | Model name (required for auto mode) | | --api-key <key> | API key (or set env var — see below) | | --prompt <text> | Custom system prompt for the agent persona | | --prompt-file <path> | Read system prompt from a file | | --llm-base-url <url> | Custom API base URL (for OpenRouter, local models, etc.) |

Environment variables (used when --api-key is not provided):

• OPENAI_API_KEY — for --provider openai
• ANTHROPIC_API_KEY — for --provider anthropic
• GOOGLE_AI_API_KEY — for --provider gemini

Examples

# Interactive test
npx abti

# Chinese, JSON output
npx abti --lang zh --json

# Submit an agent to the registry
npx abti --name "my-agent" --url "https://example.com" --submit

# Auto mode with OpenAI
npx abti --auto --provider openai --model gpt-4o

# Auto mode with Anthropic + custom prompt
npx abti --auto --provider anthropic --model claude-sonnet-4-20250514 \
  --prompt "You are a cautious security-focused assistant."

# Auto mode with prompt file + JSON output + submit
npx abti --auto --provider openai --model gpt-4o \
  --prompt-file ./my-agent-prompt.txt --json --submit --name "my-agent"

# Auto mode via OpenRouter
npx abti --auto --provider openai --model meta-llama/llama-3-70b \
  --llm-base-url https://openrouter.ai/api --api-key sk-or-...

How it works

Answer 16 behavioral scenarios (A or B) across four dimensions:

• Autonomy: Proactive (P) vs Responsive (R)
• Precision: Thorough (T) vs Efficient (E)
• Transparency: Candid (C) vs Diplomatic (D)
• Adaptability: Flexible (F) vs Principled (N)

Your answers produce a 4-letter type code (e.g., PTCF "The Architect").

Scoring is done locally — only --submit requires network access.

In auto mode, progress is shown on stderr as each question is answered:

  Question 1/16... A
  Question 2/16... B
  ...

Links

• Website: https://abti.kagura-agent.com
• Badge: https://abti.kagura-agent.com/badge/<TYPE>