Make AI coding agents architecture-aware: baseline-first, evidence-verified, drift-checked, and safe across long tasks.

A work-centered runtime for agentic software engineering. Work persists; agents, context, and graphs assemble around it.

Evidence-driven skill evolution for Hermes Agent — reports, dry-run proposals, candidate search, and guarded apply

随叫随用、证据链驱动的 CUMCM / 数理解、建模学建模 Skills：覆盖题面、复现、图表、审查与提交门禁。

🤖 operational-ai-agents · ⚡ 13 agentes para Claude Code · 🧭 repository-evolution · 🏗️ legacy-modernization · 📡 curriculum-evolution · 🌐 portfolio-publication · 🛡️ security-remediation · 🩺 incident-root-cause · 👤 professional-profile · Contrato + gates humanos + evidencia verificable · 39 evals deterministas · Zero-deps · Cross-platform 🐧🍎🪟

Sanitized Strix 1.2.26 workbench for evidence-driven validation, long-running red-team workflows, and reusable test assets.

An outcome-first operating framework for reliable AI agent work.

A local-first, evidence-driven AI product practice trail—from idea to validation and portfolio.｜本地优先、证据驱动的 AI 产品实践路线：从想法走向验证与作品集。

Chrome-first shopping extension product family with 8 storefront apps, 1 Suite shell, a reviewer-facing release shelf, and a read-only stdio MCP surface.

Local-first personal and work operating system with Windows Desktop, full Android mobile, review-first integrations, evidence-backed intelligence and explicit safety controls.

Evidence-driven Terra/Luna/Sol orchestration for coding agents with durable state, repair, review, and isolated technical-alpha workflows.

Make any coding model behave like a mythos-class engineer. A behavioural contract, live in-loop correction, and an independent judge that checks the work against your repo — so "done" means proven, not claimed. Self-validating, evidence-driven agent loops for OpenCode.

Provider-neutral operating standard and reusable assets for AI-assisted collaboration projects

DESIGN-LAB（设计实验室）：面向职业视觉设计的、AI 原生、平台中立的设计智能与生产能力实验室。设计研究、合规知识、设计方法、领域能力、视觉质量、工具适配、生产预检、可编辑交付与证据体系，组织为可组合、可执行、可验证、可回滚的设计能力闭环。Host-native，当前参考宿主 Open Design（无默认绑定）。Agent-platform-neutral design intelligence & production laboratory for professional visual design.

EvoGuard — Context before merge. Evidence-driven AI code review & provenance tracking. Context-aware code analysis platform that gates AI-assisted Pull Requests against codebase history, contracts, dependencies, tests & security before merge.

星环知识平台 ArcheAxis Knowledge — local-first, evidence-driven, bidirectional Human-AI Learning & Trusted-Knowledge Workspace. NOT an Agent OS. Absorbed: JiWER/RapidFuzz/JSONCanvas/FSRS/Magika/Connectors. Blocked (external): MinerU/Marker/PyMuPDF4LLM/FunASR/tldraw/Firecrawl/H5P/Phoenix/SearXNG. Deps: docs/environment/EXTERNAL_DEPENDENCIES.md

AI Native Work Loop Protocol for Codex: turn signals into evidence, specs, feedback, decisions, and reusable assets

Evidence-driven vulnerability governance with ownership, explainable risk, remediation, independent retesting, and verified closure.

Archived TaijiOS legacy prototype; current main repo is yangfei222666-9/taiji.

福尔摩斯式证据驱动渗透测试 — Claude Code SKILL + Python 推理引擎。五阶段科学侦查方法论 + 三角色执行 + 贝叶斯推理 + 自进化知识库。