Create Your Skill Card

A Skill Card is a set of structured files that any AI agent can discover, read, and act on. You own your data, you set the rules, and agents work for you.

Install the CLI

Install the Scoutica CLI on your machine.

macOS / Linux

curl -fsSL https://raw.githubusercontent.com/traylinx/scoutica-protocol/main/install.sh | bash

╔═══════════════════════════════════════════════════════╗
║                                                       ║
║   ⚡ Scoutica Protocol — CLI Installer               ║
║                                                       ║
║   Your skills. Your rules. Your data.                 ║
║                                                       ║
╚═══════════════════════════════════════════════════════╝

→ Creating directories in /Users/sebastian/.scoutica...
→ Downloading scoutica CLI...
→ Downloading JSON schemas...
→ Downloading card templates...
→ Downloading AI card generator...
→ Downloading validation tool...

╔═══════════════════════════════════════════════════════╗
║                                                       ║
║   ✅ Scoutica CLI installed successfully!             ║
║                                                       ║
╚═══════════════════════════════════════════════════════╝

  To get started, run:

  source /Users/sebastian/.zshrc && scoutica init

After installation, reload your shell:

source ~/.zshrc   # or source ~/.bashrc

Windows (PowerShell)

irm https://raw.githubusercontent.com/traylinx/scoutica-protocol/main/install.ps1 | iex

Note

Windows installs PowerShell implementation 0.1.0 for protocol 0.4.0, capability set windows-subset-v1. It supports init, init --ai, validate, publish, info, help, and version. The help sample below and all other commands are the POSIX implementation 0.4.0 surface.

The POSIX full help surface looks like this; on Windows, scoutica help shows only the seven commands in windows-subset-v1:

scoutica help

Scoutica Protocol CLI v0.4.0
Your skills. Your rules. Your data.

Usage:  scoutica <command> [options] [directory]

🚀 Create:   scan · init · import aijs
🔧 Manage:   info · preview · validate · publish · resolve
🏢 Employer: org · role
🌐 Network:  evaluate · jobs · send · inbox · reply · deliver · register · identity
⚙️  System:   doctor · status · logs · update · help · version

Choose a creation method

There are four ways to create your Skill Card. Pick the one that fits your workflow.

AI scan (POSIX; recommended for devs)

Put your CV, certifications, and portfolio files in a folder, then let the CLI extract your Skill Card automatically. Extraction runs locally, but a remote-capable provider may receive the full generated prompt and document text. Review the provider before scanning sensitive files.

scoutica scan ~/my-docs/

You can also specify a provider or output directory:

scoutica scan ~/my-docs/ --with gemini     # use a specific provider
scoutica scan ~/my-docs/ --with claude     # use Claude Code
scoutica scan ~/my-docs/ --with ollama     # fully offline
scoutica scan ~/my-docs/ --output ./my-card/

Supported document formats: .md .txt .pdf .docx .json .yaml .csv .html

Note

PDF text extraction requires pdftotext. Install it with brew install poppler (macOS) or apt install poppler-utils (Linux) if PDF files are not being read.

Supported AI providers (auto-detected in this order):

Provider	CLI command
Gemini CLI	gemini
Claude Code	claude
OpenAI Codex	codex
Mistral Vibe	vibe
OpenCode	opencode
Ollama	ollama
switchAILocal	ail
OpenClaw	openclaw

Interactive wizard

Answer prompts step by step to build your card without any documents.

scoutica init ./my-card/

You can also run scoutica init in the current directory, or use --ai for AI-guided instructions:

scoutica init              # current directory
scoutica init --ai         # AI-guided mode

AI paste (no install needed)

This is the easiest path and requires no CLI installation. Suitable for non-technical users.

1. Open GENERATE_MY_CARD.md on GitHub
2. Copy the entire file contents
3. Paste it into any AI assistant — ChatGPT, Claude, Gemini, Copilot, etc.
4. Follow the conversation — the AI will interview you and generate your 4 files
5. Save the files to a GitHub repo

Alternatively, use the CLI to copy the prompt to your clipboard without running a full scan:

scoutica scan ~/my-docs/ --clipboard

Then paste the clipboard contents into any AI chat.

Clone and customize

Clone the protocol repo and edit the template files directly. Gives you full access to all protocol files.

git clone https://github.com/traylinx/scoutica-protocol.git
cp -r protocol/templates/ my-card/

Edit the files in my-card/ with your data, then validate:

python tools/validate_card.py ./my-card/

Validate your card

Check your card against the protocol schemas before publishing.

scoutica validate ./my-card/

A passing result looks like:

✅ 5 checks passed

Fix any reported errors before proceeding.

Publish to GitHub

Push your card to a GitHub repository to make it discoverable by AI agents.

scoutica publish ./my-card/

You can also chain validation and publish together:

scoutica validate && scoutica publish

Your card will be live at your GitHub repository URL. Any AI agent can now discover and evaluate your profile.

What gets created

After creation, your card folder contains these files:

my-card/
├── profile.json        # Your skills, experience, tools
├── rules.yaml          # Your rules of engagement
├── evidence.json       # Links to your public work
├── SKILL.md            # Agent entry point
├── scoutica.json       # Discovery file (auto-generated)
└── rules/              # Evaluation rule templates
    ├── evaluate-fit.md
    ├── negotiate-terms.md
    ├── verify-evidence.md
    └── request-interview.md

Tip

Run scoutica info ./my-card/ at any time to see a summary of your published card.