Skip to content

Profile

SSCodeAI's public profile, focus areas, projects, and links.

Updated View as Markdown

Profile

SSCodeAI is an open-source project that organizes AI coding workflows, agent libraries, and hands-on tutorials for Japanese IT development teams.

The main focus is moving AI agents from “a convenient assistant” to “a repeatable, explainable working procedure”. Requirements, design, implementation, review, testing, incident response, and operational handover — deliverables and checkpoints are adapted to Japanese development culture.

Focus Areas

Area What SSCodeAI publishes
AI coding workflow planning, TDD, debugging, code review, verification, pre-release checks
Japanese localization prompts / skills aligned with Japanese development culture, acceptance, audit trails, incident reporting, handover
Agent library specialist roles for requirements, design, QA, PM, support, finance, legal, sales, and more
Documentation install guides, tutorials, version differences, quality management, release notes
Open source maintenance upstream tracking, explicit documentation of Japan-market diffs, generated docs sync

Projects

  • superpowers-ja — a Japanese AI coding skill pack for Claude Code, Cursor, Codex CLI, Gemini CLI, and more.
  • agency-agents-ja — an AI specialist agent library adapted for Japanese IT development, SIer, custom development, SaaS, EC, manufacturing DX, and public sector.
  • takumi — a model-agnostic orchestration platform for verifiable agentic software delivery.
  • aiitg — a zero-trust input security layer that detects, sanitizes, labels, and gates untrusted documents before they reach an LLM.
  • ai-skill-maker — a meta-skill that turns rough intent, repository evidence, and repeatable workflows into reusable AI skills and instruction bundles.
  • ai-project-skill-maker — a meta-skill that generates durable project maintainer skills and instruction bundles.
  • skills-ja-JP — the Japanese localization of the Matt Pocock Agent Skills (sscodeai/skills).
  • decision-workbench — a six-stage decision tool that expands the option space before deciding and never picks a winner.
  • decision-space-expansion — a decision methodology distilled from decision-workbench, shipped in the Agent Skills format.
  • eli5 — a picture-first explainer generator built on large SVG diagrams and minimal text.
  • fusion-cache — an LLM API caching layer that fuses exact, semantic, and prefix-cache accounting.
  • SSCodeAI site — the documentation site that gathers docs, blog, tutorials, and generated catalogs.

Working Style

SSCodeAI’s documents prioritize practical shareability.

  • Readable in Japanese, with English technical terms alongside where helpful
  • Record not only artifacts but also assumptions, rationale, and verification steps
  • Separate upstream-derived content from Japan-market originals / adaptations
  • No personal, customer, contract, or internal-spec information in public documents
Navigation

Type to search…

↑↓ navigate↵ selectEsc close