playground

跨会话长期记忆工具集 — 5 个核心 skill (recall/capture/commit/doctor/admin)，支持 OpenViking / MCP / mem0 三后端，identity 默认 fail-closed。

6 skills · 3 commands

内容创作与发布工具集，覆盖技术公众号全流程：tech-content-writer（技术文章写作，无 AI 味 + 禁用词扫描）、tech-content-audit（发布前五大维度内容审核）、wechat-cover-html（20:9 封面图，HTML + Playwright，代码密集型首选）、wechat-cover-image（封面图 Pillow 备选方案）、md-to-html（Markdown 渲染为可发布 HTML，内置 MDNice 与开源 CSS 双主题引擎）、ali-oss（上传文件到阿里云 OSS，支持多 bucket 与默认 bucket、region 自动探测、预签名链接）、wechat-publisher（将 Markdown/HTML 文章发布到微信公众号草稿箱，多主题排版 + 封面图与正文图片自动上传 + 多账号切换）。

7 skills

产品/商业/技术/设计分析工具集：product-teardown（Principal-PM 级产品拆解，Product / Business / Technology / Strategy 四层框架，Core Loop 为权重最高的一节，双语 EN+ZH 可打印 HTML 报告）、competitor-landscape（竞品二维定位矩阵）、ai-architecture-review（AI/Agent 产品的技术架构深挖：Model / Tool / Memory / Planning / Agent / Workflow / Evaluation）、copy-ui-style（从截图/URL/代码仓库逆向提取 UI 设计系统，产出 DESIGN.md + AGENTS.md + tokens.json/css + style-dna.json，带校验与 WCAG 对比度检查）、requirement-review（按风险分级的会话级需求评审，检查旧业务影响、安全、兼容性并输出版本化 brief/handoff）、plan-explainer（按风险解释 plan，支持用户复述或知情委托后交给执行流程）。

6 skills

项目演进记录系统：随代码库流转的演进日志，每条记录回答「从什么变成什么、为什么、谁牵头」。evolution-log skill 覆盖初始化 / 记录 / 查询 / 草稿确认四条工作流；Stop hook 在每轮结束前做规则快筛，命中后把「先自评是否真为方案级变更」的指令回注给会话内模型完成二级判定（opt-in、fail-open、单会话至多一次、零外发）；记录以结构化 Markdown 存进 Git，records/ 是唯一事实来源，INDEX.md 与 records.json 可随时全量重建。

1 skill · hooks

外部 AI Agent 调度工具集（Cursor / Codex 等），目前包含 cursor-cli 子 skill。

1 skill

manji 市场核心功能：版本检测、自动更新、频率控制

1 skill

AGENTS.md as project instructions, by one option. Under claude-md-or-agents-md a project with no instruction files of its own gets its AGENTS.md files instead: a prompt.context hook asks $.fs.ancestors for every AGENTS.md and .claude/AGENTS.md from the filesystem root down to the working directory and hands them to the engine as project instruction files, which it renders, announces and withholds from agents exactly as it does CLAUDE.md, skipping any file the engine already loaded by import or link; a tool.call hook on Read attaches the ones under the project root as the engine attaches a nested CLAUDE.md. Under claude-md-and-agents-md every AGENTS.md is loaded beside CLAUDE.md. Under managed-only the project's and the person's instruction files are dropped and the organization's managed ones kept. Under claude-md the engine's walk stands alone. claude-md-or-agents-md is the default.

Real-time graph intelligence — personalized PageRank, streaming delta updates, witness-signed reasoning, and federation-distributable vectors

1 skill

Personalized AWS guidance built on patterns from 350,000+ startups. Get architecture, cost, and security recommendations matched to your stage, from day-one account setup to production-ready infrastructure. Migrate to AWS from Azure, GCP, or Heroku with resource discovery, cost estimates, and generated Terraform. Move OpenAI, Azure OpenAI, and Gemini code to Amazon Bedrock, and plan and prototype AI agents on AgentCore and other AWS runtimes. Includes AWS Activate credits eligibility, 60+ exclusive startup offers, and multi-account, multi-region support. Built by AWS Startup Solutions Architects.

13 skills · 1 command · 7 agents · 1 MCP server

Turn Common Room into your GTM copilot. Research accounts and contacts, prep for calls with attendee profiles and talking points, and draft personalized outreach across email, LinkedIn, and phone.

6 skills · 2 commands · 1 MCP server

Personal knowledge brain for your coding agent — hybrid search, synthesis, and durable cross-session memory, plus a curated brain-first skill set.

80 skills · 1 MCP server · hooks

Organic LinkedIn presence, end to end, with LinkedIn's own rules enforced in code. Orchestrator (context: fork) gates every request against User Agreement §8.2 — refusing automation, scraping, engagement pods, bulk DMs, fake identity and fabricated proof, each with a compliant substitute — then routes to profile / strategy / content / engagement / analytics. Headline and whole-profile scoring with fixes ranked by points per hour; positioning brief with a mandatory exclusion list; cadence priced against real hours with a 90-minute floor; newsletter eligibility + six-month sustainability gate; post linter blocking on engagement bait and screen-reader-hostile Unicode pseudo-bold; repurposing with a reuse ledger; capped commenting roster and template-refusing outreach; analytics that test patterns against a seeded permutation null and refuse to conclude below 10 posts. 17 stdlib tools, 15 references with per-claim confidence levels. No credentials, no API calls, nothing auto-sent. Answers discussion #934.

6 skills · 8 commands · 2 agents