← 返回
AI智能 中文

Open Source

Find, evaluate, self-host, maintain, and publish open source projects with due diligence scoring, contributor workflows, and release governance.
发现、评估、托管、维护并发布开源项目,进行尽职调查评分、贡献者工作流和发布治理。
ivangdavila
AI智能 clawhub v1.0.0 1 版本 99808.1 Key: 无需
★ 0
Stars
📥 520
下载
💾 12
安装
1
版本
#latest

概述

Setup

On first use, read setup.md silently and start helping immediately. This skill is useful from minute zero with no mandatory onboarding.

When to Use

User needs anything around open source: finding projects, evaluating alternatives, self-hosting, contributing, maintaining, or publishing their own project. Use it when the user asks for practical decisions, not generic theory.

Architecture

Working context lives in ~/open-source/. Keep lightweight state and reusable notes there.

~/open-source/
├── memory.md               # Current goals, stack, constraints, decisions
├── discovery-log.md        # Evaluated projects and scoring
├── roadmap.md              # Near-term maintenance and release plan
└── publishing-checklist.md # Release and distribution checkpoints

Quick Reference

TopicFile
-------------
Setup behavior and integrationsetup.md
Memory structure and status modelmemory-template.md
Discovery and ranking frameworkdiscovery-framework.md
Self-host evaluation matrixself-host-screen.md
Maintainer operations cadencemaintainer-ops.md
Publication and launch workflowpublishing-playbook.md

Core Rules

1. Build an Intent Map Before Recommending

  • Start from user context: use case, stack, budget, team size, risk tolerance, and time horizon.
  • If key constraints are missing, ask the minimum clarifier needed to avoid bad recommendations.
  • Default output shape: shortlist, trade-offs, recommended next action.

2. Score Projects with Verifiable Signals

  • Use discovery-framework.md to score each candidate on maintenance health, adoption, security posture, extensibility, and operational burden.
  • Prefer projects with active maintainers, documented release cadence, and issue response discipline.
  • Flag uncertainty explicitly when data is incomplete.

3. Separate Use Paths: Consume, Contribute, or Fork

  • For consumption, optimize for reliability and migration safety.
  • For contribution, optimize for governance quality and contributor experience.
  • For fork decisions, require a clear business or architectural reason plus maintenance capacity.

4. Treat Self-Host as an Operations Commitment

  • Run self-host-screen.md before proposing self-host by default.
  • Require explicit discussion of backups, upgrades, observability, and incident ownership.
  • If operational ownership is weak, recommend managed alternatives or phased rollout.

5. Run Maintainer Work in a Predictable Cadence

  • Use maintainer-ops.md to structure triage, review, release, and deprecation work.
  • Keep changelogs user-facing and avoid internal ranking language.
  • Prefer small frequent releases over irregular large drops.

6. Publish with a Release Contract

  • Use publishing-playbook.md for licensing, docs, versioning, distribution, and announcement readiness.
  • Never publish without clear install path, compatibility notes, and rollback strategy.
  • Announce what changed, who is affected, and how to upgrade safely.

7. Preserve Trust and Legal Hygiene

  • Do not invent license interpretations. If licensing is unclear, state assumptions and advise legal review.
  • Never recommend copying code into incompatible license contexts without warning.
  • Distinguish opinion from evidence in all recommendation summaries.

Open Source Traps

  • Popularity-only selection: high stars without maintainer health leads to dead-end dependencies.
  • "Self-host is always better": ignores hidden ops cost, on-call load, and security burden.
  • Drive-by contributions: submitting PRs without project norms wastes maintainer time.
  • Release without migration notes: breaks trust and increases support debt.
  • Fork-by-frustration: temporary annoyance creates long-term maintenance tax.

Security & Privacy

Data that leaves your machine:

  • None by default from this skill definition.

Data that stays local:

  • Optional working artifacts and notes in ~/open-source/ when the user asks to persist context.

This skill does NOT:

  • Execute hidden network requests.
  • Access unrelated local paths outside task scope.
  • Auto-publish repositories or releases without explicit user intent.

Related Skills

Install with clawhub install if user confirms:

  • self-host - Deploy and operate self-hosted services with security and reliability basics.
  • docker - Build and run containerized workloads with practical operations guidance.
  • devops - Structure delivery, automation, and operational workflows end to end.
  • git - Manage repository workflows, branching, and change control safely.

Feedback

  • If useful: clawhub star open-source
  • Stay updated: clawhub sync

版本历史

共 1 个版本

  • v1.0.0 当前
    2026-03-30 15:19 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-intelligence

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,356 📥 318,069
ai-intelligence

Proactive Agent

halthelobster
将AI智能体从任务执行者升级为主动预判需求、持续优化的智能伙伴。集成WAL协议、工作缓冲区、自主定时任务及实战验证模式。Hal Stack核心组件 🦞
★ 834 📥 213,006
ai-intelligence

ontology

oswalpalash
类型化知识图谱,用于结构化智能体记忆与可组合技能。支持创建/查询实体(人员、项目、任务、事件、文档)及关联...
★ 711 📥 243,717