AI & ML

repoprompt Bundle

Validate RepoPrompt CE contributions before committing or pushing. Use whenever an agent is about to create a commit, push the current branch, rewrite history, delete a branch or fork, or change GitHub-visible repository state. Enforces staged-index and outgoing-range secret scanning, repository guardrails, clean push boundaries, an explicit PR-ready lane for path-selected heavyweight validation, and explicit approval for destructive Git or visible live-app operations.

repoprompt Bundle

Apply evidence-led RepoPrompt CE maintainership principles distilled from documented project guidance. Use when planning, scoping, implementing, reviewing, triaging, or sequencing RPCE changes; deciding whether work should be an investigation, issue, PR, or follow-up; checking compatibility, migrations, performance, observability, model defaults, or release risk; or asking how a proposed change fits the project's intended direction.

ooooooooooooooooooop Bundle

通过 chatgpt-web 的 MCP 或 REST 桥，把本地 agent 接入已登录的 ChatGPT Web。 适用于“推进/继续网页端会话”“让网页端设计、本地执行”“协作/对等推进”，以及只读拉取网页端会话/项目资料和 conversation_id 续接。 桥只传递文本；网页端没有本地 shell、文件系统或其他本地工具。

ooooooooooooooooooop Bundle

在重要任务上运行完整世界模型闭环的执行协议——恢复持久状态→建立竞争模型→显式预测→真实观察→修订模型→主动取证→决策→持久化。由 AGENTS.md World Model Gate 判 OFF/CORE/FULL 后加载；用于多步、高风险、跨会话、存在竞争解释或需要持久世界模型的真实任务，不用于简单一次性任务。

ooooooooooooooooooop Bundle

对已进入 OPERATIONS MODE 的 Personal AI / DSH 基础设施做只读运维复查：聚合 governance、durability、aic drift、proposal、memory/model/routing 与 personalization 行为监控（纠正率/重复纠正/选择泄漏），区分 NO ACTION、REVIEW、ACTION REQUIRED 与 EXTERNAL BLOCKER；用于日常健康检查、异常复盘、升级前检查，或用户询问“系统现在健康吗/需要管什么/AI 最近是不是又变难用了/个性化有没有退化”时。默认不修改 canonical，不为变绿而静默修复。

64x-lunicorn Bundle

Cuts a Spec into agent-sized tickets and drafts each ticket body as a vertical slice with the Spec's Gherkin scenarios as acceptance criteria. Use when a Spec is split into tickets, or when a single ticket is written or reshaped so an agent can implement it.

64x-lunicorn Skill

Traces factual claims to the primary source that owns them and records each with source, date, version and confidence. Use when research or a design discussion rests on a fact about an API, library, standard, law or market, when a claim sounds plausible but nobody has checked it, or when reading legwork can be delegated to a background agent.

64x-lunicorn Skill

Sets up a project for the 64x-lunicorn plugin through a guided interview with detected defaults, writes the committed marker .claude/64x-lunicorn.yml that every other skill reads, and generates the CI gate, issue templates, agent docs, community files and README from it.

64x-lunicorn Bundle

Writes a project's agent docs from fixed templates, CLAUDE.md with the project's conventions, CONTEXT.md for domain terms and docs/adr/ with the decision format and a generated index, keeps recorded conventions, terms and decisions on re-runs and reports drift. Use when a project gets its agent docs for the first time, when CLAUDE.md or the ADR index drifted, or when docs from an older agent setup such as AGENTS.md or docs/agents/ are replaced.

godot-fun Skill

Converts storyboard markdown (bilingual Chinese + English narration per shot) into speech audio via IndexTTS2 (same stack as ai-text-to-speech). Batch-writes WAVs under Chinese/ and English/ named by shot id (model loaded once), pads 0.4 s edge silence in place, then a speech-timeline.md and one concatenated SRT per language (Chinese.srt / English.srt). Use when the user wants storyboard TTS, storyboard to speech, narration VO, storyboard-to-speech, bilingual VO export, subtitles, or batch TTS from a storyboard.md.

allagentsdev Skill

Manage AllAgents workspaces, plugins, skills, client targets, global profiles, and MCP servers through the allagents CLI. Use when a user asks to initialize, inspect, update, or configure AllAgents; install or remove plugins or skills; manage profiles; or add, inspect, authenticate, update, or remove MCP servers. Resolve current command syntax from machine-readable help instead of relying on remembered flags.

cannedjean Bundle

Delegate a small, self-contained coding task to the Antigravity CLI (`agy`, Google's terminal coding agent) in headless mode and collect the files it produced. Use when the user says things like "이건 agy한테 시켜", "Antigravity로 만들어와", "delegate this to agy", or wants a bounded coding chore (write a script/module/tests/boilerplate/refactor of a few files) done by a second agent and the outputs brought back. Not for tasks that need interactive approval, credentials, or multi-repo changes.

btspoony Bundle

Use when a Morning Star agent must choose, run, or interpret `mstar-harness` / `mstar` CLI commands — picking the command family for a task (plan row, lifecycle close, register write, report landing, worktree and lease checks, path resolution), satisfying the preconditions a family needs (harness root, control root, neutral cwd, session envelope, version tokens), reading exit codes 0 / 1 / 2 and stable refusal codes, or running the versioned read-modify-write handshake. Also load it when a topical skill's engine-check callout points at the CLI, or when a command refused and the cause looks like a missing precondition. Flags and option wording are never restated here — the CLI help owns them.

wkentaro Bundle

Launch a task in a new Herdr tab after choosing the agent harness and model.

xy040427-collab Bundle

Create Agent4PPT full-slide-image and editable-mode PowerPoint decks and paired comparisons. For every editable page, generate a complete slide first, erase selected text with an image model, then restore native text through PowerPoint MCP and visually match the original. Includes the portable runtime, stage records and review tools.

kerpopule Skill

Use on passages a search just returned (memory, vault, session history, wiki, web) before reading them in. Jev ranks them, drops the irrelevant, and flags prompt injection hidden in the text.

kajitiluna Bundle

Create and edit ERD Designer .erd files (database entity-relationship diagrams). Use when the user asks to design database tables, add or modify tables, columns, relations, indexes or memos in a .erd file, to export DDL from a .erd file, or to check whether a .erd design matches another revision or a live database.

oyadotai Skill

Control a real browser via Oya Browser MCP tools. Use this skill when the user asks you to browse the web, interact with websites, fill forms, scrape data, or automate browser tasks. Provides the mental model and strategies for effective browser automation through structured markdown and numbered element IDs.

wowyuarm Bundle

Create, install, revise, or retire this Member's private skills, including how a skill manages its configuration and credentials. Read this skill before creating, installing, revising, or retiring any skill in this Member's private skills directory, or when repeated or fragile work has produced a reusable method worth preserving as a skill.

btspoony Bundle

Morning Star (晨星) harness 计划目录约定 —— `{HARNESS_DIR}` / `{PLAN_DIR}` / `{SDD_DIR}` / `{ITERATION_DIR}` / `{KNOWLEDGE_DIR}` / `{SPECS_DIR}` / `{WORKFLOW_DIR}` / `{PROJECT_DIR}` 发现与初始化（默认 `.mstar/`，兼容 `.agents/`）、`docs` 与 harness 子树边界、review bundle、未启用 plan 时的工作方式、Spec 集成分支与多 Plan 实现分支（显式 base / merge 靶 / PR target）、Morning Star plan-writing path gate、工期预估（agent-oriented）。**必须**在读写 `.mstar/` / `.agents/`、初始化 harness、编排含 plan 的任务、或对齐 `metadata.primary_spec` 时 Read；`@project-manager` 开 plan 任务前必读。plan 文件 / status / residual / review bundle / knowledge → **`mstar-artifacts`**；分支与 QC 检出 → **`mstar-branch-worktree`**。

btspoony Bundle

Agent skill 撰写 / 重写 / 优化规范（SkillsBench 实验门控）。在新建任意 skill、大改 SKILL.md、改写 description 触发契约、压缩过长 body、评审 skill 结构、或判断某段行为塑形文案是否值得保留时读取。适用于任何领域的 skill，不限于本仓库专题；不用于普通应用实现。

btspoony Bundle

[Cursor maint] Morning Star 路由与 prompt 迭代评估 —— 维护/回归专用（非日常 PM 编排 runtime）。路径：本仓库 `.cursor/skills/mstar-routing-eval/`。(1) PM 路由回归：`assets/routing-evals.json`；(2) Prompt/规则迭代评估；(3) Routing Eval Report 模板。变更 PM 路由或 harness 门禁、`@prompt-engineer` 改 agents/skills、或显式跑场景集时 Read。

kerpopule Skill

Use to pick the cheapest model that is good enough for a turn — choosing a model, delegating a sub-task to a sub-agent, cutting model spend, or setting up and tuning Jev routing pools.

wowyuarm Skill

Maintain this Member's private memory — what earns a line in the memory.md index, what belongs in a notes/ file instead, how to compact an index past half its budget, and when to demote knowledge that no longer earns a line. Read this skill when the injected memory block shows a maintenance warning, when the index approaches its budget, or before adding or renaming a note.

huaweicloud Skill

Clean up HuaweiCloud DevKit environment for fresh testing. Removes plugin files, skills, MCP server, KooCLI, and Codex plugin registration. Use when the user wants a clean test environment or to reset before reinstall.

huaweicloud Skill

Clean up HuaweiCloud DevKit from a CodeArts Agent (.codeartsdoer) environment for fresh testing. Removes codearts skills, MCP registration, plugin runtime dir, KooCLI (hcloud dirs + PATH entry), and restores the default sandbox mode. Use when the user wants to fully reset the CodeArts Agent environment before reinstalling huaweicloud-devkit --target codearts.

huaweicloud Bundle

每晚自动回归测试 huaweicloud-devkit 插件的指引能力。使用条件：需要本地凭证文件、需要真实华为云账号、任务为插件能力回归验证时。

chadparker Bundle

Build and sharpen a project's domain model. Use when discussing codebase terminology, writing or editing a CONTEXT.md, or recording or editing an ADR.

heronbo111 Bundle

生成各类 AI 视频的提示词：①图生视频（图片/首尾帧/多图驱动，含道具与主体替换）②文生视频（纯文字）③参考视频生视频（参考运镜/动作/节奏）④参考视频替换生视频（保留动作构图、只换主体）；也用于拆解 AI 二创/爆款成片（读片三问 + 机器证据 + 拆解报告）。当用户要求写视频提示词、把文案/台词/素材变成生成指令、做分镜、换人换物换装、复刻或仿做某条片子，或在即梦/Seedance/MiniMax H3 等平台生成视频时使用。限额按模型维护，见 `references/platforms.md` 的「模型与限额」表；本 skill 只产出提示词文字与分析报告，不提交生成、不消耗积分；并根据每次成片反馈持续优化规则。

microsoft Skill

Update the shadow knowledge base after code changes and from conversational insights. Detects what changed via git diff, refreshes per-file shadows, captures knowledge shared by the user during the session, and updates cross-cutting discoveries. Invoke manually with /shadow-frog-update; the preToolUse hook will remind the agent when the shadow is behind HEAD.

justtyashh Skill

Captures meaningful behavioral events from agent sessions without altering agent responses. Produces BehavioralTrace artifacts. Use during Soulcatcher observation phase, when analyzing session history, or when the user asks to watch and record how they work.

rojim666 Bundle

Analyze PDF files by converting them into agent-readable text, OCR text, page render images, and extracted embedded images without polluting the working directory. Use when an agent needs to read, summarize, inspect, compare, quote, or reason about PDFs, especially when the foundation model cannot ingest PDFs directly or when tables, figures, scans, screenshots, or layout matter.

justtyashh Skill

Compiles an accepted ProcedureDraft into a portable Agent Skills SKILL.md plus evidence file in learned-skills/. Use after Soulcatcher challenge-inference accepts a procedure and the user approves compilation.

justtyashh Skill

Converts recurring patterns from PatternReport data into candidate executable procedures with separated evidence and inference. Produces ProcedureDraft artifacts. Use during Soulcatcher inference phase or when the user asks what workflow their behavior suggests.

justtyashh Skill

Adversarially tests ProcedureDraft hypotheses before compilation — tries to disprove patterns using counterevidence and context checks. Use during Soulcatcher validation phase or when the user asks to verify a learned workflow is real.

rojim666 Bundle

Use when the user asks how to build with OpenAI products or APIs, asks about Codex itself or choosing Codex surfaces, needs up-to-date official documentation with citations, help choosing the latest model for a use case, latest/current/default-model prompting guidance, or model upgrade and prompt-upgrade guidance; use OpenAI docs MCP tools for non-Codex docs questions, use the Codex manual helper first for broad Codex self-knowledge, and restrict fallback browsing to official OpenAI domains.