seed-forge/harness-ai-kit
Package manager for AI agent assets — 42 skills, 5 CLIs, 1 plugin. Skills for AI/LLM agent engineering, eval-driven dev, spec-driven dev, database (MySQL/PG/Redis/Kafka/Mongo/Oracle/NL2SQL), K8s/Docker diagnostics, infra ops (Dify/Nexus/Harbor/SonarQube) & docs/patent. Runtimes: Codex, Claude Code, Cursor, Kiro, DSH.
项目介绍Project Overview
harness-ai-kit 是 AI 代理资产的包管理器,统一安装、锁定并治理跨 Codex、Claude Code、Cursor、Kiro 的技能、CLI、MCP 与循环,附带企业级 AI-ops 技能库。核心是基于 manifest 与 lockfile 的组合式管理,能复用并编排现有社区资产。适合需要跨运行时一致治理 AI 资产、避免环境漂移的团队。需注意当前仍为 v0.1,循环框架与注册中心尚未发布。
harness-ai-kit is a package manager for AI agent assets that installs, locks, and governs skills, CLIs, MCPs, and loops across Codex, Claude Code, Cursor, and Kiro, bundling an enterprise AI-ops skill library. Its core capability is composition-first asset orchestration via a versioned manifest and SHA-256 lockfile, reusing existing community tools rather than replacing them. Use it when teams need consistent, auditable AI-asset governance across multiple runtimes. Caveat: the project is at v0.1, so loop automation and the public registry are not yet released.
请帮我了解并安装插件:【harness-ai-kit】【https://github.com/seed-forge/harness-ai-kit】
把上面这条消息直接发给当前会话里的 DSH,让它帮你了解并安装。安装命令不一定准确,发给 DSH 更稳。Send this message to DSH in your current session. CLI install commands may not be accurate across systems — DSH will figure it out for you.
或使用命令行安装(适合开发者)Or use CLI install (for developers)
命令行安装CLI Install
dsh plugin --profile web add github:seed-forge/harness-ai-kit
把 seed-forge/harness-ai-kit 加入你的 DSH 配置(web profile)即可启用。
READMEREADME
harness-ai-kit
Package manager for AI agent assets (skills / CLIs / MCPs / plugins / loops).
Install skills from any GitHub repository into Codex, Claude Code, Cursor, Kiro or DeepSeek Harness (dsh) with dependency resolution, lockfiles, checksums and rollback.
First-class dsh runtime
harness-ai-kit install skill <id> --runtime dsh --scope projectinstalls into.agents/skills(dsh rank 200 native scan, zero config).harness-ai-kit install skill <id> --runtime dsh --scope globalinstalls into~/.agents/skills(dsh rank 500 native scan).harness-ai-kit doctor dshchecks the dsh baseline (@deepseek-ai/dsh@0.1.0-rc.6, pnpm >=10,DSH_HOME, profile dirs).harness-ai-kit install plugin harness-ai-kit-plugin --profile webinstalls the bundled dsh plugin (session tools + virtual skill provider; zero workspace bloat).
See docs/dsh-integration.md for the full integration model.
Deterministic test boundary
Asset authoring and test policy is bundled in
docs/asset-authoring-contract.md. Unit and combination tests are offline and
use temporary fixtures/mocked transports. Real registry, Nexus, TLS, credential
and role-bound publishing checks belong to an explicit integration run:
$env:HARNESS_AI_KIT_INTEGRATION = "1"
python -m pytest -q --run-integration -m integration
The default test command must remain reproducible without external services:
python -m pytest -q
- Repository: https://github.com/seed-forge/harness-ai-kit
- License: Apache-2.0
pip install --upgrade harness-ai-kit==0.18.2
harness-ai-kit init
harness-ai-kit add skill https://github.com/OWNER/REPO/tree/main/path/to/skill
harness-ai-kit sync
nexu-io/open-design
ruvnet/ruflo
amruthpillai/reactive-resume
esengine/DeepSeek-Reasonix
volcengine/OpenViking
Molunerfinn/PicGo
titanwings/distilly
titanwings/colleague-skill