Skip to main content

​
Syntax

scoutica init [options] [directory]

​
Options

​
directory
string
Target directory for the card files. Defaults to the current directory.
​
--ai
flag
Skip the interactive wizard and instead print AI-guided instructions for creating your card using any AI assistant (ChatGPT, Claude, Gemini, Copilot, etc.).

​
Usage examples

# Create card in the current directory
scoutica init

# Create card in a specific folder
scoutica init ./my-card

# Get AI-guided creation instructions
scoutica init --ai

# AI mode with a specific output folder
scoutica init --ai ./my-card

​
Interactive wizard flow

When run without --ai, the wizard walks you through six sections:
1

Identity

Enter your professional name, job title, seniority level (entry → executive), years of experience, and current availability status.
2

Skills and expertise

Enter your primary domains, key skills, tools and platforms, certifications, and specializations — all as comma-separated lists.
3

Languages

Add spoken languages with proficiency levels (native, fluent, professional, basic). You can add as many as you need.
4

Summary

Enter your education background and a 2–3 sentence professional summary.
5

Rules of Engagement

Define which engagement types you accept (permanent, contract, fractional, advisory, internship), your minimum compensation, remote policy, blocked industries, and preferred tech stack keywords for matching.
6

Public evidence

Add links to public work that proves your skills — GitHub repos, portfolios, certificates, articles, case studies, videos, and more.

​
Files created

After the wizard completes, the following files are written to the target directory:
FileDescription
profile.jsonYour skills, experience, domains, and availability
rules.yamlYour Rules of Engagement — compensation floors, remote policy, auto-reject rules
evidence.jsonLinks to verifiable public work
SKILL.mdAgent entry point — how AI agents should read and use your card
rules/Four evaluation rule templates (evaluate-fit, negotiate-terms, verify-evidence, request-interview)
.gitignorePre-configured to keep secrets out of git

​
AI mode (--ai)

The --ai flag prints step-by-step instructions for using any AI assistant to generate your card through conversation. The CLI will offer to copy GENERATE_MY_CARD.md to your clipboard so you can paste it directly into ChatGPT, Claude, Gemini, or Copilot. 
scoutica init --ai

  The easiest way to create your card with AI:

  1. Copy the contents of this file:
     ~/.scoutica/GENERATE_MY_CARD.md

  2. Paste it into any AI assistant:
     ChatGPT, Claude, Gemini, Copilot — any will work

  3. Follow the conversation — the AI will interview you
     and generate all 4 files

  4. Save the generated files to: ./my-card/
When --ai is selected, no files are written. The wizard is skipped entirely. Use this mode if you prefer a conversational AI interface over answering CLI prompts.

​
Security notes

  • All file generation uses python3 json.dump / yaml.dump — never raw shell variable interpolation — preventing injection attacks.
  • Before writing, the CLI checks whether any target path is a symlink and refuses to overwrite it, preventing symlink attacks.
  • The generated .gitignore helps prevent accidental commits of sensitive files.

​
Next steps

After running init, validate and publish your card: 
scoutica validate ./my-card
scoutica publish ./my-card