#
ENJA

compass

tabiya-tech/compass

Compass is an AI-powered platform designed to help job-seekers discover, articulate, and catalog their skills using the ESCO taxonomy and inclusive livelihood frameworks.

14 13MITUpdated 2026-06-29

Overview

This project serves as an AI conversational agent system mapping user work experiences to standardized occupations and skills. It features a Python/FastAPI backend utilizing Google Vertex AI, a React/TypeScript frontend chat application, and Pulumi-based GCP infrastructure to provide an inclusive skill-discovery tool for emerging markets and informal economies.

Capabilities

  • Conversational skill extraction
  • ESCO taxonomy mapping
  • Structured output generation via Pydantic and Vertex AI
  • Multi-language localization support
  • Cloud Run deployment via Pulumi IaC

Best for

Mapping informal and formal work experience to standardized ESCO skill taxonomies, Populating digital skill wallets for job-seekers in emerging markets, Guiding conversational interviews to extract unrecognized competencies and soft skills

Works with

GitHub Copilot
Evoa Score breakdown= Σ (score × weight)
69
Task usefulness20%88 → +17.6

実タスクにどれだけ役立つか(機能の豊富さ・用途の明確さ)。 AIによるcapabilities/use-cases解析

Code quality15%90 → +13.5

実装・指示の品質。 AIによるSKILL.md/README解析

Maintenance15%85 → +12.8

リポジトリがどれだけ活発に保守されているか。 GitHub 最終push日時の新しさ

Documentation12%34 → +4.1

ドキュメントの充実度・分かりやすさ。 README/独自要約の情報量

Security15%50 → +7.5

危険・不審な挙動が無いか。 AIによるセキュリティレビュー

Originality10%85 → +8.5

ありふれたラッパーではない独自性。 AIによる独自性判定

Popularity8%40 → +3.2

コミュニティの採用度。 GitHub Stars/Forks(対数スケール)

Compatibility5%25 → +1.3

対応AIエージェントの広さ。 AIによる対応エージェント判定

Weighted total68.4 / 100

ライセンス不明/制限あり(Red)のSkillは総合スコアに0.85倍の補正を適用します。 ランキングはこのScoreのみで決まり、広告で変わりません。 算出方法の詳細 →

Security considerations

Handles sensitive personal data; requires strict adherence to data protection guidelines and environment variable segregation.

Categories

Summary and analysis are original content generated by AI Skills Rank. The skill's source text is not reproduced here — view it on the linked repository.