Skip to content
View sscodeai's full-sized avatar
🎯
Focusing
🎯
Focusing
6 followers · 0 following

Achievements

Achievement: YOLOAchievement: Pair Extraordinairex2Achievement: Pull Shark

Achievements

Achievement: YOLOAchievement: Pair Extraordinairex2Achievement: Pull Shark

Block or report sscodeai

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
sscodeai/README.md
SS

👋 Hi, I'm SS (sscodeai)

Agentic SWE / AI delivery platform engineer in Japan.

I build AI that must survive production: affordable, trustworthy, accountable. My work is organized around one belief — AI entering enterprise production has to pass three gates at once, and I build a system for each gate.

Frugal — 95% of requests should never wake a big model. Safe — distrust the input, distrust the keys. Verified — Agents propose. Humans decide.

🔐 Safe — keys & inputs, distrust by default

Project What it does
aiitg AI Input Trust Gateway — hidden-content auditor for documents fed to LLMs/agents (zero-width chars, white text, hidden sheets). 102 tests.
keysmith Go MCP server + CLI for agent-safe secrets: age-encrypted, masked views, self-healing rotation. Plaintext never enters agent context.

💰 Frugal — every token spent where it counts

Project What it does
fusion-cache LLM API caching middleware: exact-match → semantic → upstream prefix-cache. 91 tests, up to 97% cost cut.
dsh-llm-cache LLM response cache merged upstream into DeepSeek Harness (@deepseek-ai/dsh-llm-cache) — my contribution to the official repo.

✅ Verified — agents propose, humans decide

Project What it does
takumi Model-agnostic platform for verifiable agentic software delivery: Manager/Executor/Auditor + verify-fail → repair → re-verify. 87/87 tests; 23-task eval 96% first-pass → 100% final.
proctor Agent Eval tool: score, audit & explain agent trajectories — metrics + causal failure attribution.
decision-workbench Expand the decision space before you decide — multi-model committee + human final call. Hidden-gold eval: 26% → 78% → 100% (12/12).
decision-space-expansion The methodology behind decision-workbench, as a portable Agent Skill (EN/JA). Works with Claude Code, Cursor, Codex, Gemini CLI.

🌏 Japanese-market work

Project What it does
agency-agents-ja 323 production-ready AI agents for Japanese dev teams — Claude Code / Cursor / Copilot.
skills-ja-JP · superpowers-ja Japanese localizations of engineering skill packs for AI coding agents.

Sites:

  • www.sscodeai.com — my portfolio: story, resume & project deep-dives
  • sscodeai.com — docs system (Nimbus, Cloudflare open-source framework): product docs, tutorials & Japanese-market agent toolkits

Pinned Loading

  1. agency-agents-ja agency-agents-ja Public

    323個の即戦力AI専門エージェント — Claude Code/Cursor/Copilot等に対応。SIer/受託開発/SaaS/製造DX/公共分野など日本市場向け114個のオリジナルを含む、agency-agents日本語コミュニティ版

    Shell 4

  2. takumi takumi Public

    Model-agnostic platform for verifiable agentic software delivery: workflows, traceability, evidence-native testing, agent evals, and verify-before-done loops.

    TypeScript

  3. keysmith keysmith Public

    The keysmith for AI agents: Go MCP server + CLI for agent-safe secrets with age encryption, masked views, handle passing, self-healing rotation, and Vault short-TTL credentials.

    Go

  4. aiitg aiitg Public

    AI Input Trust Gateway — hidden content auditor for documents fed to LLMs/Agents

    Python

  5. ai-skill-maker ai-skill-maker Public

    AI Skill Maker is a meta-skill for creating and refreshing reusable AI skills and assistant instruction bundles across platforms.

    JavaScript 1

  6. proctor proctor Public

    Agent Eval tool: score, audit & explain agent trajectories — metrics + causal failure attribution in pure Go.

    Go