知项 · Knowject

让项目知识，真正为团队所用。

Project-anchored Skills for Claude Code and Codex. Turn PRDs, design mocks, API specs, and repository context into concrete handoff artifacts.

简体中文 · Skills · Documentation · Contributing · Security

What Knowject Is

Knowject now starts from Skills.

It is a source-available Skill pack and AI workspace for teams that want AI agents to work from real project context instead of generic prompts. The first usable surface is skills/: a set of installable, project-grounded Skills that Claude Code and Codex can consume directly.

The goal is simple: make handoff work cheaper. A PRD should become a usable HTML mock. A design screenshot should become a component decomposition plan. An OpenAPI file should become a typed client scaffold. A project should carry its own context, source-cited memory, and review artifacts so the next agent run does not start from zero.

What Knowject Skills Do

Input	Skill	Output
Existing project repository	knowject-context-init	knowject/context.yaml with stack, brand, API, and design paths
Written PRD	knowject-prd-to-mock	Brand-aware single-page HTML mock
UI screenshot, PDF page, or Figma export	knowject-read-design	Component decomposition plan and skeleton files
Express routes or OpenAPI document	knowject-read-api	Endpoint inventory and typed-client scaffold
OpenAPI 3 document	knowject-api-to-types	TypeScript response types wired into the generated client
Project artifacts, handoff notes, diffs, or decisions	knowject-memory-capture	Source-cited durable project memory under knowject/memory/
Project knowledge sources	knowject-rag-eval	Source-cited RAG and citation eval cases under knowject/evals/
API surfaces or endpoint inventories	knowject-mcp-tool-designer	MCP tool design artifacts under knowject/tools/

These Skills are intentionally not generic brainstorming or code-review prompts. Each one is scoped to a cross-role handoff where project context matters.

How It Works

1. Run knowject-context-init once in your project.
2. Commit knowject/context.yaml as the project anchor.
3. Optionally run knowject-memory-capture to preserve source-cited project facts, decisions, workflows, risks, and lessons under knowject/memory/.
4. Use the other Skills to turn PRDs, UI mocks, API specs, knowledge sources, and endpoint inventories into concrete review artifacts.

knowject/context.yaml is the shared contract. It records the project type, frontend and backend stack, design source paths, API source paths, client output directory, and brand tokens. It must not contain secrets, base URLs, API keys, or environment-specific config.

your-project/
  knowject/
    context.yaml
    README.md
    memory/
      README.md
      project-memory.yaml
    evals/
      README.md
      rag-eval-cases.yaml
      rag-eval-report.md
    tools/
      README.md
      mcp-tools.plan.md
      mcp-tools.schema.json
      tool-risk-report.md

Install

Recommended: ask your AI agent

Copy this prompt into Claude Code or Codex:

根据指引安装 Knowject AI toolkit：https://github.com/lynxlangya/knowject/blob/main/skills/install.md

The agent will read skills/install.md, use a stable local checkout, run the existing installer, and verify the Claude Code + Codex Skill links.

Manual fallback

Requirements

• Claude Code or Codex
• macOS / Linux, or Windows with WSL
• bash, python3 3.8+, git
• Python PyYAML for full validation

git clone https://github.com/lynxlangya/knowject.git
cd knowject
bash skills/scripts/install.sh

install.sh itself does not require PyYAML. Full validation via skills/scripts/verify.sh does:

pip install pyyaml
bash skills/scripts/verify.sh

The installer symlinks every knowject-* Skill plus the shared support folder _shared into:

~/.claude/skills/
~/.codex/skills/

Restart Claude Code or Codex after installation.

Use

In any project:

/knowject init

or:

帮我在这个项目里启用 knowject

After that, use the Skills directly:

/knowject mock
/knowject design
/knowject api
/knowject types
/knowject memory
/knowject rag eval
/knowject mcp tools

Natural-language requests work too, as long as the intent matches the Skill:

把这段 PRD 出一个高保真 HTML 原型。
把这张设计稿拆成 React + Antd 组件骨架。
找一下用户列表 endpoint，并生成 typed client。
把这个 OpenAPI 文档转成 TypeScript response types。
把这次交接总结成项目记忆。
基于这些知识库文档生成 RAG 引用评测用例。
把这些 endpoint 设计成带风险 gate 的 MCP tools。

Skill Boundaries

Skill	Good for	Deliberate limits
knowject-context-init	Project onboarding and context detection	Does not guess secrets or environment config
knowject-prd-to-mock	Static HTML mock from a written requirement	Does not build runtime interactivity
knowject-read-design	UI decomposition and component skeletons	Does not wire state, routes, tests, or real data
knowject-read-api	API discovery and typed-client scaffold	Phase 2 supports Express and OpenAPI sources
knowject-api-to-types	OpenAPI 3 response types for generated clients	Day-1 scope is response types, not full request typing
knowject-memory-capture	Source-cited file-based project memory	No DB, vector store, platform UI, daemon, secrets, or uncited facts
knowject-rag-eval	Source-cited RAG and citation eval artifacts	No live model calls, Chroma, MongoDB, platform UI, network scoring, or external eval dependency
knowject-mcp-tool-designer	API-to-MCP tool design with risk labels and confirmation gates	Does not implement an MCP server, runtime handlers, auth code, API routes, or UI

This makes the output reviewable. A Skill should produce a useful first pass, not silently mutate a production codebase.

Validate the Skill Pack

bash skills/scripts/verify.sh
bash skills/scripts/test-install.sh

verify.sh checks the manifest, Skill frontmatter, Codex adapters, per-Skill README coverage, context.yaml examples, route extractors, OpenAPI extractors, typed-client generation, brand extraction, framework extraction, type extraction, client rewrite fixtures, the memory validator, RAG eval validator, and MCP tool schema fixture. test-install.sh checks installer idempotency and installed shared-file references.

The Platform Behind the Skills

Knowject also contains the product workspace that these Skills can grow into. That platform is already more than a shell:

• authenticated product shell with project routes, global asset pages, members, and settings
• Express API modules for auth, projects, members, knowledge, skills, agents, and settings
• project chat read/write flow with SSE streaming, replay/edit, source seeds, and citation patch events
• global and project-private knowledge flows with upload, diagnostics, retry, rebuild, and search
• structured Skill governance with binding validation and project-chat Skill injection
• Python indexing runtime for parsing, chunking, embedding, and Chroma orchestration
• Docker-based local and production-style deployment baselines

The Skill pack is the lead surface. The workspace is the longer-term system for managing project memory, team Skills, knowledge assets, and project chat.

Repository Layout

skills/        Installable Knowject Skills for Claude Code and Codex
apps/
  platform/    React frontend product shell
  api/         Express business API
  indexer-py/  Python indexing runtime
packages/
  request/     Shared HTTP client package
  ui/          Shared UI components
docs/          Project documentation source of truth
.agents/       Internal project-governance Skills
.codex/        Project-local Codex configuration

Tech Stack

• Skills: Markdown Skill specs, Codex adapters, Python validation scripts
• Frontend: React 19, Vite 7, Ant Design 6, Tailwind CSS 4
• API: Express 4, TypeScript, MongoDB Node.js Driver
• Indexing: Python 3.12+, uv, Chroma
• Auth: JWT + argon2id
• Tooling: pnpm workspace, Turborepo, ESLint, Prettier
• Infra: Docker Compose, MongoDB, Chroma, Caddy

Development Quick Start

Requirements

• Node.js >= 22
• pnpm 10
• Python 3.12+
• uv

Host workflow

cp .env.example .env.local
pnpm install
pnpm dev

pnpm dev starts platform + api + indexer-py through the workspace.

Docker-managed dependencies

pnpm dev:init
pnpm dev:up

Useful commands

pnpm dev:web
pnpm dev:api
pnpm test
pnpm check-types
pnpm build
pnpm verify:global-assets-foundation
pnpm verify:index-ops-project-consumption
pnpm verify:core-loop-readiness
pnpm docker:local:health
pnpm knowject:help

Documentation

docs/ is the documentation source of truth. docs/exports/ is a derived export layer and not the primary fact surface.

• Skills README
• AI Install Guide
• Skills Schema
• Project Rules
• Documentation Index
• Current Architecture
• Project Chat Sources
• Skills Governance
• Contracts Index
• Chat Contract
• Skills Contract
• Platform README
• API README
• Docker README

Status

• Shipped Skill pack: knowject-context-init, knowject-read-api, knowject-prd-to-mock, knowject-read-design, knowject-api-to-types
• Current package version: 0.1.0
• Next distribution step: Claude Code plugin manifest and marketplace listing

Contributing

Contributions are welcome. Start with CONTRIBUTING.md for setup, workflow expectations, validation rules, and documentation sync requirements. For new Skills, follow skills/_shared/contributing-skills.md.

Security

For vulnerability reporting and current support scope, see SECURITY.md.

License

This repository is currently distributed under the Knowject Proprietary Source-Available License.

Personal, non-commercial learning, private study, evaluation, and non-production experimentation are allowed. Commercial use, company use, client use, deployment, hosting, SaaS, distribution, or monetized derivative work requires the Licensor's prior written permission.