AI & ML

rt22766 Bundle

通过 reasoning_effort、Magic String、组合推理题和离线日期题快速检测当前 API 是否为真实 Claude 模型，并在需要时升级到身份、工具、元数据与嵌套层级的深度审查。用于怀疑模型真假、来源异常、被第三方包装，或需要输出模型真实性检测报告时。

alexmeckes Bundle

Persistent `godot-mcp` and AI Bridge workflows for Godot 4.x projects. Use this skill for `.tscn`, `.gd`, `.gdshader`, `.tres`, and live editor tasks that need scene inspection, iterative node edits, refresh/run/test loops, runtime diagnostics, or hybrid file-plus-editor validation.

aquifer-sea Bundle

Constrains the Agent to follow a structured bug fix process — locate → root cause → fix → regression verify.

aquifer-sea Bundle

Constrains the Agent to follow a safe refactoring process — guarantee functional equivalence while improving code quality.

aquifer-sea Bundle

约束 Agent 必须遵循结构化的 Bug 修复流程：定位 → 根因分析 → 实施修复 → 回归测试验证。

aquifer-sea Bundle

约束 Agent 必须遵循安全的重构流程：在保证原有业务逻辑和功能完全等价的前提下，提升代码质量。

aquifer-sea Bundle

Constrains the Agent to perform structured code reviews across correctness, security, performance, and maintainability.

aquifer-sea Bundle

Constrains the Agent to follow a structured feature development process — requirements → design → implement → verify.

aquifer-sea Bundle

约束 Agent 从正确性、安全性、性能和可维护性四个维度进行系统化的结构性代码审查。

aquifer-sea Bundle

约束 Agent 必须遵循结构化的功能特性开发流程：理解需求 → 技术设想 → 编写代码 → 自验自测。

youzaiagi Bundle

Spawn coding agents (codex/claude/gemini) in parallel across isolated git worktrees for async background execution. Use when user mentions "run in background", "parallel tasks", "spawn agents", "worktree tasks", "async execution", or wants to work on multiple features simultaneously. Fire-and-forget by default - spawn tasks and return immediately.

alubiama Skill

Critically refine user requests into stronger, safer, and more executable task specs. Use whenever a user shares an idea, plan, requirement, or prompt. Detect bottlenecks, reinvented solutions, blind spots, cognitive biases, missing constraints, unclear success criteria, weak sourcing, and unknown-unknown risks; then rewrite and request user confirmation before execution.

berrzebb Bundle

Session orchestrator for the consensus-loop — reads handoff, picks unblocked tasks, distributes to parallel workers, tracks agent assignments, manages correction cycles via SendMessage. Use when starting a work session with pending tasks, distributing implementation work, or reviewing completed worker output.

berrzebb Bundle

Run consensus-loop deterministic analysis tools via CLI — code_map, dependency_graph, audit_scan, coverage_map, rtm_parse, rtm_merge, audit_history, fvm_generate, fvm_validate. Use this skill whenever you need codebase analysis (symbol index, dependency DAG, pattern scan, coverage), RTM operations (parse, merge, query), or FVM operations (generate matrix, validate against server) — even if the MCP server is not configured. This skill replaces mcp__plugin_consensus-loop_consensus-loop__* tool calls with equivalent CLI commands.

salihcantekin Bundle

Prompt Optimizer Copilot Skill

ubranch Skill

Use this skill whenever the user wants to parallelize work, orchestrate multiple agents, use agent teams, or says things like "do it with orchestration", "write an agent team", "make it parallel", "orchestrate this", "create agents for this", "multi-agent", or asks for a task that would benefit from being split into parallel work streams. Also triggers when user pastes a task and wants it done faster/better with multiple agents. Works for any domain — coding, research, design, testing, auditing, debugging, writing, analysis.

salihcantekin Bundle

Prompt Engineering Copilot Skill

salihcantekin Bundle

Agent Workflow Designer Copilot Skill

alpha-park Skill

Use this skill when the user wants to search across multiple AI models and web sources simultaneously using GenPark's aggregated search. Triggers: 'search genpark', 'ask genpark', 'use genpark to find', 'compare AI answers on', 'multi-model search'. NOT for: product shopping on GenPark (use genpark-shop), or standard web searches the user wants from a single source.

harshamendu Bundle

Model Layer

c7a6 Bundle

Use when the agent must act as a cron worker for pending jobs from this MCP server, including polling work, following or inspecting workflow guidance, renewing leases, and reporting completion or failure. Do not use when the task is to define cron jobs, edit repository code, or answer general questions without processing queued cron work. Success means the agent either completes or fails the claimed run with the correct lease token and workflow metadata, or explicitly reports that no actionable cron work is available.

alpha-park Skill

Use this skill when the user wants to publish, announce, or share an OpenClaw skill on Moltbook so other agents can discover and use it. Trigger phrases include "publish my skill to Moltbook", "share this skill on Moltbook", "announce my skill on Moltbook", "post my skill to the agent community", "submit skill to Moltbook", "let other agents use my skill", "share genpark skill", or any request to make a locally-written skill available to the wider agent community.

atheerium Bundle

Meta-skill for creating workflow-optimized OpenCode skills from MCP servers. Use when you need to build a new skill that integrates local or remote MCP tools into a specialized, context-efficient workflow.

palmdr Skill

Delegates a task to OpenAI Codex CLI as a subagent and returns its response. Use this skill whenever the user wants Codex's opinion, a second opinion from Codex, wants to ask Codex to do something, wants to compare Claude vs Codex on a task, or says anything like "ask Codex", "have Codex look at this", "what does Codex think", "run this through Codex", or "get Codex to review". Also use when the user wants multiple AI perspectives on code, writing, or analysis.

palmdr Skill

Delegates a task to Google Gemini CLI as a subagent and returns its response. Use this skill whenever the user wants Gemini's opinion, a second opinion from Gemini, wants to ask Gemini to do something, wants to compare Claude vs Gemini on a task, or says anything like "ask Gemini", "have Gemini look at this", "what does Gemini think", "run this through Gemini", or "get Gemini to review". Also use when the user wants multiple AI perspectives on code, writing, research, or analysis — Gemini has a very large context window useful for long documents.

palmdr Skill

Self-evolving scanner that discovers trending agentic systems, Claude Code configurations, and AI agent patterns, then proposes concrete improvements to the user's own Claude Code setup. Runs 5 parallel scan agents (GitHub trending, community discussions, registry updates, deep repo analysis, setup audit), synthesizes findings, writes a dated report, updates the tools registry, and proposes specific CLAUDE.md improvements as diffs. Use this for "scan for agentic trends", "run agentic radar", "what's new in Claude Code / AI agents", "audit my setup", or weekly automated runs via launchd.

koyonx Bundle

Use this skill when the user wants to start from a reusable prompt template for review, refactoring, tests, or a custom template stored under ~/.codex/prompt-templates.

gitstq Skill

Intelligent context management system for AI coding agents. Solves the critical problem of context window overflow by implementing a tiered context architecture with smart prioritization, automatic summarization, and dynamic loading. Features include: automatic context budget calculation based on model capacity, smart file prioritization using dependency graph analysis, conversation summarization with key decision preservation, and a rolling context window that keeps the most relevant information always available. Use when working on large codebases, long conversations, or when the AI agent starts "forgetting" earlier context. Trigger keywords: context management, context overflow, 智能上下文, 上下文管理, conversation summary, long conversation, large codebase, context window.

gitstq Skill

Self-improving AI agent evolution engine with intelligent experience management. Captures learnings, errors, and feature requests from every interaction, deduplicates them intelligently using semantic similarity, and promotes high-value patterns into reusable skills. Unlike basic memory skills, it features automatic context compaction, cross-project knowledge transfer, team collaboration patterns, and a built-in review cycle. Works across all sessions without any external service. Use when you want your AI agent to learn from mistakes, remember your preferences, accumulate project knowledge over time, or continuously improve its responses. Trigger keywords: self-improve, learn, remember, evolve, 自我改进, 记住经验, 学习模式, continuous improvement, memory management.

berrzebb Skill

Run a consensus-loop audit manually — Codex reviews pending trigger_tag items in the watch file. Use when you want to trigger an audit without editing the watch file, re-run a failed audit, or test the audit prompt.

berrzebb Skill

Show current consensus-loop status — pending reviews, audit state, retro marker, agent assignments. Use to check what's happening before starting work or after returning from a break.

xcrrr Skill

Use this skill when crafting, iterating, or optimizing prompts for LLMs including zero-shot, few-shot, chain-of-thought, role prompting, structured output, and prompt chaining. Not for fine-tuning or training models. Not for evaluating model quality across benchmarks.

xcrrr Skill

Use this skill when comparing AI or LLM models on benchmarks, capability, cost, latency, context window, or task-specific fit to help teams select the right model for their use case and budget. Not for training or fine-tuning models. Not for building eval frameworks from scratch.

jamestorrevillas Skill

MCP Builder

jamestorrevillas Skill

AI Agent Design

maksimlokhmakov Bundle

Firecrawl MCP Server integration with browser automation for OpenClaw. Provides web scraping, crawling, search, LLM extraction and interactive browser automation.