AI & ML

testdino-hq Skill

Review code for bugs, MCP protocol violations, consistency issues, and convention adherence

testdino-hq Skill

Scaffold and implement a new MCP tool end-to-end — endpoint, handler, registration, docs

testdino-hq Skill

Debug a malfunctioning MCP tool — trace the full request lifecycle to find the issue

rapid-recovery-agency-inc Bundle

Use when deciding which database MCP tool to call, routing between platform vs tenant databases, looking up tenant or company info, or any task referencing routing tables (platform_companies, client, server) or tenant data tables (address_*, case_*, hotspot_*, lpr_*, drn_*).

joaquincampo Bundle

Use when the user asks to consult the advisor, wants transcript-aware strategic guidance, wants a second opinion on ongoing agent work, provides a transcript for review, is stuck, is choosing between candidates, or wants to know whether completed work is ready. Always run the advisor as a separate GPT 5.6 Sol xhigh agent with the exact recovered advisor system prompt.

joaquincampo Skill

Use when spawning Codex subagents or Codex threads and choosing among GPT 5.6 Luna, Terra, and Sol with the appropriate effort. Use this with orchestrating-subagents whenever delegation involves model overrides, reasoning effort, Explore agents, checker/fixer loops, or fan-out work.

t1nker-1220 Skill

Expert systematic debugging through root cause analysis. Reflects on 5-7 possible sources, distills to 1-2 most likely causes, adds strategic logging to validate assumptions, and confirms diagnosis before fixing. Uses MCP servers for database validation. Use when troubleshooting errors, debugging issues, investigating bugs, analyzing problems, or when user mentions error, bug, not working, broken, or debugging.

jwmyers Bundle

This skill should be used when working on the Zero-Day Attack Unity project. Provides guidance for coordinating specialized agents across ALL development activities including bug investigation, implementation, design exploration, architecture decisions, Unity/MCP integration, and multi-phase planning. Trigger phrases: "fix this bug", "implement", "design", "investigate", "refactor", "review", "plan", "coordinate", "which agent", "help me with", "explore the codebase", or any development work on this Unity game project.

itsmostafa Skill

Patterns and architectures for building AI agents and workflows with LLMs. Use when designing systems that involve tool use, multi-step reasoning, autonomous decision-making, or orchestration of LLM-driven tasks.

itsmostafa Bundle

Building and training neural networks with PyTorch. Use when implementing deep learning models, training loops, data pipelines, model optimization with torch.compile, distributed training, or deploying PyTorch models.

itsmostafa Skill

Crafting effective prompts for LLMs. Use when designing prompts, improving output quality, structuring complex instructions, or debugging poor model responses.

itsmostafa Bundle

Strategies for managing LLM context windows effectively in AI agents. Use when building agents that handle long conversations, multi-step tasks, tool orchestration, or need to maintain coherence across extended interactions.

mqfacultyofarts Bundle

Run, wait for, diagnose, and report PromptGrimoire CI lanes submitted to the shared bunyip wrapper. Use for bunyip act, nightly, browser, NiceGUI, quality, image-build, or other heavy-lane work where an agent must track completion, inspect failures, verify cleanup, or hand back exact evidence.

mqfacultyofarts Skill

Run a production incident investigation end-to-end — collect telemetry, build dataset, generate review report, write postmortem. Operationalises the incident-analysis methodology skill with concrete commands for PromptGrimoire's infrastructure.

kecbigmt Skill

Execute codex CLI as sub-agent for thorough code reviews (codex's specialty), alternative approaches when stuck, or complex problems requiring extended investigation. Note that inference takes significant time.

bauhaus-infau Bundle

Convert GHX files to LLM-readable format and generate RAG summaries. Use when users mention GHX, Grasshopper definition, .ghx files, convert for LLM, document Grasshopper, or RAG context for parametric workflows.

dyai2025 Skill

Unterstützt dabei, aus vagen oder konkreten Vorhaben einen realistischen Projektplan mit Spezifikation, Task-Breakdown und passenden LLM-Prompts für Umsetzungsagenten zu entwickeln; verwenden, wenn Planung, Dokumentation und Prompt-Design für ein neues Vorhaben nötig sind.

zealousear Bundle

Universal LLM Router — route prompts to any model across all providers

zealousear Bundle

Multi-model reasoning protocol for Claude Code that runs five independent solver drafts (across different LLM providers), four adversarial debate reviews, and a reliability-weighted aggregation step to select or synthesize the final answer. Invoked via /debate slash command.

surrealdb Skill

Reach what the company brain's files no longer say — a superseded version of a risk, a relationship nobody wrote down, context filed by a session you never saw — using the brain_recall tool. Use it only when that tool is present; it is absent unless agent memory is configured, and then the `brain` skill alone is the whole answer. Read the files first with `brain`, then come here for the why and the history behind them.

hollaugo Skill

Run automated tests on your ChatGPT App using MCP Inspector and golden prompt validation.

hollaugo Skill

Add a new MCP tool to your ChatGPT App. Guides through tool design, schema creation, and code generation.

ai-dynamo Bundle

Use when reviewing AIPerf code for the SEMANTIC qualities that mechanical linters cannot catch — error-message informativeness, type-hint descriptiveness, docstring example usefulness, naming disambiguation, convention explicitness, comment density, and reference-file exemplariness. Invoke after mechanical checks (`make check-ergonomics`, `make check-ruff-baselined`) are green, before shipping a branch or PR. Explicitly out of scope are correctness bugs, style (tabs/quotes), preferences, and anything the guardrails already catch.

victorvvedtion Skill

The Quant God. Data over narrative, systematic edge, probability thinking, never override the model. Archetype: quant.

patchy631 Bundle

Add and manage evaluation results in Hugging Face model cards. Supports extracting eval tables from README content, importing scores from Artificial Analysis API, and running custom model evaluations with vLLM/lighteval. Works with the model-index metadata format.

kirilltrubitsyn Bundle

Комплексный аудит безопасности приложений, API, репозиториев, AI/agent-систем, MCP-интеграций, инфраструктуры и цепочки поставок. Использовать при явном запросе — «аудит безопасности», «security audit», «проверь на уязвимости», «security review», «threat model», «hardening», «проверка перед продакшеном», «prompt injection», «MCP security», — а также при secure-design review или планировании авторизованного пентеста. Не использовать для обычного code review, отладки или рефакторинга без запроса о безопасности.

antorsae Skill

[TMUX MODE] Send code to Codex via tmux file-based IPC. Only use when user explicitly runs /codex-review command. For natural language requests like 'review with codex', use the delegate_codex_review MCP tool instead.

antorsae Skill

[TMUX MODE] Delegate implementation to Codex via tmux file-based IPC. Only use when user explicitly runs /codex-implement command. For natural language requests, use the delegate_codex_implement MCP tool instead.

antorsae Skill

Delegate arbitrary implementation, debugging, refactoring, planning, review, or custom coding tasks to ChatGPT GPT-5.5 Pro through the chatgpt-pro-worker MCP server. Use when the user asks to use ChatGPT Pro, GPT-5.5 Pro, or a Pro subagent/worker for a task.

vukhanhtruong Bundle

Create or update professional architecture documentation through a structured interview, live research, and code scanning. Use when the user asks for architecture docs, system documentation, an architecture review doc, C4 diagrams, ADRs, a threat model, or to document an existing codebase.

dopsonbr Bundle

Create new Agent Skills with proper structure, templates, and best practices. Use when building custom skills for Claude Code, GitHub Copilot, or other Agent Skills-compatible tools.

dopsonbr Bundle

Validate Agent Skills against best practices and specification requirements. Use when reviewing skills, before publishing, or to audit existing skills.

ddnetters Skill

LLM observability with Langfuse — query traces, generations, costs, metrics, and debug LLM pipelines via the REST API

simoncai519 Bundle

Use when user wants to work with Shopify Admin API, Storefront API, validate Liquid code, explore GraphQL schemas, build Shopify apps, or inspect Shopify documentation and MCP tools.

z1993 Bundle

Use when the user wants to route, choose, learn, or apply Dennett-style intuition pumps and thinking tools for hard problems, stuck reasoning, argument testing, concept confusion, weak explanations, or agent self-rescue after repeated unproductive reasoning.

z1993 Bundle

全自动生成前沿技术领域的全景图谱。适用于以技术为核心的赛道（如量子计算、核聚变、AI Agent 等），通过双向挖掘（正向追踪学术源头 + 逆向溯源独角兽创始团队）构建学术→开源→商业化的完整脉络。触发词：「技术mapping」「技术全景图」「technology map」「赛道图谱」。