Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.scoutica.com/llms.txt

Use this file to discover all available pages before exploring further.

The protocol transcends singular job applications. By adhering to rigorous schema data parsing, a Scoutica Skill Card acts as an autonomous ambassador mapping perfectly across disparate professional domains.

​
1. The Stealth Mode Engineer

The Problem: You work on top-secret infrastructure at a major cloud provider. You don’t want a noisy LinkedIn profile explicitly defining your exact tasks, but you still want recruiters who operate within rigorous parameters to contact you if they hit your $250k Base floor. The Execution:
  • You instantiate purely via scoutica init, completely skipping the --ai engine parsing because your true data must be isolated.
  • In your rules.yaml, you configure the auto_reject block aggressively, filtering out explicitly defined industries (e.g., ["crypto", "ads", "gambling"]).
  • You compile your evidence.json referencing open-source repos completely unrelated to your corporate identity to prove baseline competency without breaching NDAs.
  • When an AI agent reaches out dynamically, your card mathematically blocks the evaluation instantly if their parameters misalign, leaving your inbox spotless.

​
2. The Remote Open-Source Contractor

The Problem: You do high-level independent consulting logic. Your hours are entirely asynchronous and strictly remote. The Execution:
  • You run scoutica scan ~/GitHub-Projects/ ./ allowing your local Mistral/Gemini LLM directly read your heavy architectural markdown overviews and structure your capability matrix completely.
  • In rules.yaml, you forcefully assert engagement.allowed_types: ["contract", "freelance"] and set remote.policy: "remote".
  • By pushing your profile back to GitHub and issuing scoutica preview, you generate an instant brutalist HTML application you link explicitly into your Twitter bio _Discover my skill API constraint limits._

​
3. The Autonomous Headhunter AI

The Problem: As a recruiter, you waste 35 hours weekly manually parsing PDFs and guessing if candidates will accept an exact salary band or require visa sponsorships. The Execution:
  • Your AI pipeline leverages Scoutica directly from the backend.
  • It parses candidate URLs aggressively via scoutica resolve <card-url>.
  • Without using token-heavy LLMs for extraction, standard Python libraries index rules.compensation.minimum_base_eur and immediately drop candidates matching less than the target budget array computationally.
  • You engage in interviews exclusively with candidates programmatically capable of clearing Phase 1 matching metrics intrinsically via structured JSON filtering.
The applications branch fundamentally into autonomous ecosystems precisely because the metadata is rigorous and validated. You are no longer managing PDFs; you are managing a living database schema.