AI & ML

jesse-black Bundle

Create, update, revise, reopen, or rescope a plan or ExecPlan using `PLANS.md`. Use when an agent needs to create a plan, plan a change, update the plan with new requirements, or rewrite an active ExecPlan.

jesse-black Skill

Review the current branch, PR, or worktree against the plan or ExecPlan and record findings. Use when an agent needs a code review with findings for changes associated with an active plan.

jesse-black Bundle

Implement from an active plan or ExecPlan while keeping it up to date. Use when an agent needs to implement the active plan, continue plan-driven work, resume the current plan, or carry out an existing ExecPlan.

jggomez Skill

Guides development, structured formatting, and iterative refinement of multimodal prompts for the Gemini Omni world model. Use this skill when a user requires composition or modifications of video, text rendering, multi-input syncing, camera direction, or style morphing.

jggomez Skill

Skill for Gemini Omni Flash (gemini-omni-flash-preview) video prompt engineering best practices. Use when the prompt_architect agent needs to structure or refine a video generation prompt, apply the 6-dimension framework, or enforce preservation guardrails on edit turns. Activates when user requests involve video generation, editing, style changes, or camera adjustments.

jlevy Skill

Git-native issue tracking (beads), coding guidelines, knowledge injection, and spec-driven planning for AI agents. Drop-in replacement for bd/Beads with simpler architecture. Use for: tracking issues/beads with dependencies, creating bugs/features/tasks, planning specs, implementing features from specs, code reviews, committing code, creating PRs, loading coding guidelines (TypeScript, Python, TDD, golden testing, Convex, monorepo patterns), code cleanup, research briefs, architecture docs, agent handoffs, external tracker and Linear integration setup, personal Linear API key onboarding, viewing beads in a live browser, and checking out third-party library source code. Invoke when user mentions: tbd, beads, bd, shortcuts, issues, bugs, tasks, features, epics, todo, tracking, specs, planning, implementation, validation, guidelines, templates, commit, PR, pull request, code review, testing, TDD, test-driven, golden testing, snapshot testing, TypeScript, Python, Convex, monorepo, cleanup, dead code, refactor, ha

jlevy Bundle

Fast, consistent Markdown auto-formatter for typographic cleanup, normalization, and clean semantic line breaks. Use when creating, editing, or cleaning Markdown; formatting LLM-generated docs; adopting Flowmark in a repository; adding Markdown format scripts or commit hooks; or replacing Prettier or another Markdown formatter.

jumodada Skill

Use when testing an authorized Cloudflare Turnstile integration or operating an authorized production challenge with drissionpage-mcp. Covers official visible, invisible, pass, fail, and forced-interactive test keys; fresh iframe geometry; bounded coordinate interaction; refresh recovery; parent-page postconditions; screenshots; and token-safe evidence.

jumodada Skill

Use when a drissionpage-mcp task targets an iframe such as a payment widget, challenge, SSO flow, or embedded checkout. Classifies frame document access and outer presentation evidence, then selects a DOM, viewport-coordinate, scroll, keyboard, or parent-page verification path.

justrach Skill

Add, connect, or invoke an MCP server in this workspace — use when the user says "using the MCP skill", asks to add/connect/enable/turn on an MCP, pastes an MCP URL or npx package, or wants DeepWiki/Mobbin tools this session.

kentcdodds Skill

Orchestrate sub-agents for large tasks inside a single environment: plan, delegate coding to cheap/fast models, parallelize the critical path, keep reviews lean, and close every loop. Use when acting as an orchestrator or writing a kickoff for one. Prefer implement when fan-out does not clearly pay. For multi-environment fleets, use conduct instead.

langwatch Skill

Set up comprehensive evaluations for your AI agent with LangWatch — experiments (batch testing), evaluators (scoring functions), datasets, online evaluation (production monitoring), and guardrails (real-time blocking). Supports both code (SDK) and platform (CLI) approaches. Use when the user wants to evaluate, test, benchmark, monitor, or safeguard their agent.

ledgerhq Skill

Gitmoji + Conventional Commits convention for this repo (commit messages, branch names, and PR titles), enforced by the Danger CI bot. Use whenever creating commits, invoking /commit, naming a branch or PR, or fixing messages that failed Danger.

ledgerhq Skill

End-to-end release workflow to cut and publish a new version (consume version plans, bump versions, update changelogs, open the release PR against main). Use whenever the user asks to do a release, cut a version, publish packages, or invokes /release. Do not improvise the steps — follow this procedure exactly.

ledgerhq Skill

Gitflow backmerge workflow that brings release changes from main back into develop. Use whenever the user says "backmerge", "back-merge", "merge main into develop", or invokes /backmerge. Do not improvise the steps — follow this procedure exactly.

liranmao Bundle

Build submission-grade manuscript figures in Python or R from one shared style spec, for bioinformatics data (single-cell, spatial omics, histology, model benchmarks). Use when creating, restyling, or QA-ing a figure or panel for a paper, thesis, or preprint - including UMAP/embedding plots, expression heatmaps, image plates, forest/effect plots, and multi-panel composites. Triggers on "figure", "panel", "plot for the paper", "publication quality", "Nature figure", "journal figure", "figure legend", "export figure", "colorblind safe", "300/600 dpi".

mmo66666 Bundle

Use when a designer provides a UE5 storyboard, shot list, timed visual script, character-action beat, camera intent, or requests Sequencer, CineCamera, VibeUE, or MCP execution, validation, diagnosis, or rework for game CG, trailers, cutscenes, showcases, or ads.

mmo66666 Bundle

面向设计师的 UE5 地编统筹 Skill。基于 Unreal Engine 5.8 官方 MCP 与可选 VibeUE，在编辑器中规划、白盒搭建、改造和验收关卡；覆盖现状审计、尺度与动线、地形、水体、道路、POI、植被/PCG、场景布置、World Partition、可玩性、截图和性能检查。用户提到地编、关卡设计、关卡白盒、地图搭建、开放世界、场景布置、地形生态、路线和 POI、地图优化，或要求 AI 通过 MCP/VibeUE 操作 UE 关卡时使用。

machinesoul11 Skill

Separate project enthusiasm from business reality when a user wants to monetize an idea they've already decided to build. Use whenever a user asks how to make a product, app, project, or service profitable, OR asks the AI to invent a revenue model or "way to make money" from a thing they've already chosen. Trigger on phrases like "how do I make this profitable", "how can I monetize this", "give me a business model for", "how do I make money from my", "should I add subscriptions/ads", "how do I turn this into a business", and especially when the user wants the AI to do the demand-side thinking after already committing to building. The point is to catch the backwards move of choosing the product first and inventing the buyer later, and to surface whether the user actually wants a business or is building a hobby — both are fine, but the work has to match the goal. Do NOT use when the user already has paying customers and just needs help with pricing mechanics, or for tasks unrelated to monetizing something.

maguroid Bundle

Fork the current Codex context into a side agent only on explicit $agent-fork, /agent-fork, or named invocation; not ordinary delegation.

mathruffian-dot Skill

收工同步助手（三層級自動偵測）。當使用者說「收工」、「結束了」、「下班」、「準備換電腦」、「同步」、「先到這裡」、「換電腦繼續做」等任何要結束工作並保存進度的請求時，請一定要使用此技能。本技能會更新 AGENTS.md 進度與 handoff.md 交接檔、git commit + push、把詳細紀錄寫進 Obsidian，確保下次（或在另一台電腦、或換一個 Agent）打開能無縫接續。

mlevison Bundle

Stress-test a claim, decision or conclusion one question at a time, using Paul and Elder's Elements of Reasoning and Intellectual Standards. Use when the user says "critical thinking", "check my reasoning", "poke holes in this", "is this a good argument", "am I missing something", "how do I know this is true", or wants their thinking tested before acting on it. If the text came out of a model, hand off to critical-thinking-genai.

mlevison Bundle

Interrogate model-generated text with eleven probes that test the output itself, because there is no author to walk back through the reasoning. Use when the user says "is this AI answer any good", "did it make this up", "check what ChatGPT told me", "can I trust this", "verify this AI output", "hallucination", or pastes model output they are about to rely on. Also use when an answer earlier in this conversation turns out to be wrong, before correcting it in place.

naren-m Skill

Analyze a codebase and recommend Cursor automations (hooks, subagents, skills, plugins, MCP servers).

nekuda-ai Bundle

Add WebMCP tools to a website codebase so browser agents can act through the site's own logic instead of scraping the page. Use when asked to make a site agent-ready, add WebMCP or document.modelContext tools, or implement site tools with @nekuda/webmcp-sdk.

nsharandroidnstudio Skill

Use when an agent must run long-horizon, open-ended, parallelizable tasks (deep research, large multi-file code changes, hundreds of record lookups) and a single-loop agent drifts after 5-20 steps. Encodes the four-pillar deep-agents pattern — planning tool (write_todos), virtual filesystem for context offload, isolated subagents, long-term memory — plus the token/KV-cache economics that decide whether it's viable and a step-count rule for when NOT to use it. Reach for it past ~15-20 steps with separable sub-tasks; stay single-loop below that.

nsharandroidnstudio Skill

Use when designing, building, or refactoring tools (function calls / MCP tools) that an AI agent will invoke — how to consolidate tools around workflows instead of wrapping API endpoints, namespace them, return high-signal natural-language context instead of raw IDs/UUIDs, offer concise vs detailed response formats, stay token-efficient with pagination/filtering/truncation, write actionable error messages, prompt-engineer tool descriptions, and drive the whole thing with evaluation-driven development. Reach for it whenever an agent misuses tools, calls too many, chokes on raw tool output, or you're standing up a new MCP server.

phasespace-labs Bundle

Search, save, and manage persistent agent memory using Palinode (git-versioned markdown + vector search). Use when user says 'remember this', 'save to memory', 'what do you know about X', 'search memory for context', 'what was decided about', 'register a trigger', 'show memory diff', 'run consolidation'. Also fires when an orchestrating agent needs to: retrieve prior context before a task, save a decision for future sessions, check what is known about an entity, or surface relevant memory before proceeding. Output: structured JSON list with content, category, score fields. Do NOT use for Mem0 queries, session-only notes, or direct file browsing.

phasespace-labs Skill

Guide intentional persistent-memory capture during coding sessions via Palinode MCP. Recall at session start; save durable milestones, explicit wrap-ups, and user-requested memories. Do NOT write for recall-only/no-new-information sessions or after an explicit no-save request.

phasespace-labs Bundle

Use Palinode persistent memory in Claude Code via MCP (13 tools for search, save, git diff, triggers, consolidation). Use when user says 'save this decision', 'remember why we chose X', 'search memory before we start', 'what do we know about this project', 'show what changed in memory', 'register a trigger for this topic', 'set up Palinode MCP'. Also fires when an agent needs to: load prior context at session start, persist an architectural decision, retrieve past decisions before writing code, or surface relevant memory before proceeding with a task. Output: structured JSON. Do NOT use for Mem0, session-only notes, or running the Palinode API server.

professorpalmer Bundle

Operate and supervise Puppetmaster through MCP or CLI. Use for non-trivial edits, implementations, audits, reviews, broad investigations, CodeGraph lookups, routing decisions, long-running start_* jobs, MCP disconnects, stuck/empty/degraded jobs, and any request to monitor or recover Puppetmaster work.

romankurnovskii Bundle

Validate Etemaro strategy JSON files through the Etemaro CLI and report whether each strategy is valid or invalid, plus which fields are unknown and will therefore be ignored by the runtime. Use this skill whenever the user asks to validate, lint, check, or sanity-check a strategy JSON file (config/shared/strategy-library.json, an agent strategy entry, or a pasted strategy object), asks "is this strategy valid?", "why is my strategy field ignored?", "check the strategy schema", or "does this strategy have unknown fields?". Trigger even without the word "validate": pointing at a strategy JSON and asking what is wrong or what will be used is this skill.

ryanwaite Bundle

Rewrite text in the Classic Style described by Steven Pinker in "The Sense of Style" (drawn from Thomas & Turner's "Clear and Simple as the Truth"). Strips AI writing artifacts (em-dashes, "not just X but Y", delve/leverage, rule-of-three padding, reflexive hedging) and rewrites prose as a clear window onto the world: concrete nouns, active verbs, a visible agent, no metadiscourse. Use when the user says "rewrite in classic style", "classic style this", "Pinker-ify", "de-AI this", "make this sound human", or asks to remove em-dashes and AI tells.

sakamoto-family-smile Bundle

Web research assistant skill using Brave Search MCP

sakamoto-family-smile Bundle

Skill: hotcook-recipes

dmno-dev Skill

Varlock Docs Search