AI & ML

ericm1018 Skill

Use when a user asks about AI health checkups, token usage, LLM cost or spend, OpenAI Anthropic DeepSeek Qwen usage, BYOK vault setup, provider API keys, slow agents, context cleanup, tool cleanup, cost optimization, or installing and activating SkillFM Beacon MCP.

phperkael Bundle

PostgreSQL read-only data inspection skill. Use when users ask an agent to inspect rows, verify business state, compare owner and derived tables, or produce safe SELECT queries without running DML, DDL, or repairs.

phperkael Bundle

Nacos configuration operation skill with guarded read, create, update, delete, snapshot, publish, and runtime verification workflow. Use when users ask an agent to inspect or change Nacos dataIds while preserving rollback evidence and redacting secrets.

errahmounispace-copilot Bundle

Triage issues through a state machine driven by triage roles. Use when user wants to create an issue, triage issues, review incoming bugs or feature requests, prepare issues for an AFK agent, or manage issue workflow.

errahmounispace-copilot Bundle

Build a throwaway prototype to flesh out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".

luokai0 Bundle

This skill should be used when the user asks to "implement LLM-as-judge", "compare model outputs", "create evaluation rubrics", "mitigate evaluation bias", or mentions direct scoring, pairwise comparison, position bias, evaluation pipelines, or automated quality assessment.

luokai0 Bundle

This skill should be used when the user asks to "compress context", "summarize conversation history", "implement compaction", "reduce token usage", or mentions context compression, structured summarization, tokens-per-task optimization, or long-running agent sessions exceeding context limits.

luokai0 Bundle

This skill should be used when the user asks to "diagnose context problems", "fix lost-in-middle issues", "debug agent failures", "understand context poisoning", or mentions context degradation, attention patterns, context clash, context confusion, or agent performance degradation. Provides patterns for recognizing and mitigating context failures.

luokai0 Bundle

This skill should be used when the user asks to "understand context", "explain context windows", "design agent architecture", "debug context issues", "optimize context usage", or discusses context components, attention mechanics, progressive disclosure, or context budgeting. Provides foundational understanding of context engineering for AI agent systems.

luokai0 Bundle

This skill should be used when the user asks to "optimize context", "reduce token costs", "improve context efficiency", "implement KV-cache optimization", "partition context", or mentions context limits, observation masking, context budgeting, or extending effective context capacity.

luokai0 Bundle

End-to-end GitHub repository maintenance for open-source projects. Use when asked to triage issues, review PRs, analyze contributor activity, generate maintenance reports, or maintain a repository. Triggers include "triage", "maintain", "review PRs", "analyze issues", "repo maintenance", "what needs attention", "open source maintenance", or any request to understand and act on GitHub issues/PRs. Supports human-in-the-loop workflows with persistent memory across sessions.

luokai0 Bundle

Knowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."

chb-learner Skill

用户说"调研XXX领域的文章"、"帮我搜索XXX相关论文"、"我想了解一下XXX方向"时使用。 根据用户输入的主题，用 LLM 生成多样化检索关键词，然后自动抓取、点评、更新 summary 候选列表。 调研阶段不精读论文，不调用 paper-reader，不下载 PDF，不抽图。 **触发词**：调研、搜索、了解、看看XX方向、XX领域文章

luokai0 Bundle

Postgres performance optimization and best practices from Supabase. Use this skill when writing, reviewing, or optimizing Postgres queries, schema designs, or database configurations.

elijj Bundle

Use this skill when the user asks to grade, review, score, improve, rewrite, harden, sanity-check, or optimize a Codex `/goal` prompt or any long-running autonomous Codex objective. Trigger for phrases like "grade this /goal", "is this a good goal", "optimize my goal", "score my autonomous task", "goal grader", or when a prompt starts with `/goal` and the user wants feedback before running it. The skill grades execution readiness, identifies blockers, and returns a safer optimized `/goal` rewrite.

whiner9 Skill

1C:Naparnik (1С:Напарник) MCP server tools - BSL code analysis, review, rewriting, ITS knowledge base, platform documentation search, configuration documentation. Use when checking BSL code quality, searching ITS standards, reading platform docs, comparing versions, or finding configuration-specific documentation.

elijj Bundle

Use this skill whenever the user wants to build, convert, test, benchmark, package, or improve a Codex skill. It applies to SKILL.md structure, frontmatter, trigger wording, bundled scripts/references/assets, agents/openai.yaml metadata, .skill packaging, trigger evals, baseline comparisons, review viewers, forward-testing, and iterative description optimization. Skip it for adjacent tasks like Python/npm packaging, GitHub Actions, generic model evals, plugin creation, README writing, or standalone scripts unless they are explicitly part of a Codex skill.

elijj Bundle

Use this skill whenever writing, editing, reviewing, debugging, or refactoring code and the work could be over-scoped, ambiguous, or risky. This is the default coding-behavior layer for non-trivial work: make surgical changes, surface assumptions, avoid speculative abstractions, preserve existing style, and define verifiable success criteria. Trigger it for feature work, bug fixes, refactors, reviews, and any prompt asking for 'simple', 'minimal', 'don't overdo it', 'surgical', or 'Karpathy guidelines'.

elijj Bundle

Use this skill only when the user wants to convert, migrate, port, translate, adapt, or get parity for a skill, prompt, command, agent, hook, rule, MCP config, or workflow from another agent harness into a Codex skill. This applies to Claude Code skills and slash commands, Cursor/Cline/Roo/Aider rules, prompt packs, and AGENTS/CLAUDE instructions. Trigger for phrases like "make this work in Codex", "full parity", "convert this skill", or "bring over this Claude workflow". Do not use for native Codex planning, PRP, TDD, build-fix, code review, PR, commit, E2E, packaging, or docs tasks unless the user is porting them from another harness.

whiner9 Skill

1C:Enterprise (1С:Предприятие) development tools via EDT MCP server — BSL code analysis, metadata inspection, module navigation, error checking, debugging, configuration management. Use when working with 1C code, BSL modules, 1C metadata, 1C queries, 1C forms, or any 1C:Enterprise development task.

playerrch Skill

Generate architecture, workflow, ER, and module diagrams for the graduation thesis "单一课程场景下智能教学 Agent 的模块化设计与实践" using a unified, advisor-approved visual style. Use this skill whenever the user asks to "画一张图", "重新出图", "画架构图", "更新图片", "改图", "redraw figure", or otherwise requests any figure that will end up in the thesis or its revision packages (`docs/论文修改内容包_*.docx`, `docs/论文图片更新包_*.docx`, `docs/饶纯含--毕业论文_*.docx`). The skill encodes every figure rule the advisor and the student have signed off on so the agent does not have to re-discover them.

playerrch Skill

Forces a lightweight README reconnaissance before any broad code exploration or multi-file modification in an unfamiliar repository. Use at the start of every new conversation on a project the agent has not yet read, whenever the user asks an open-ended question like "this project does what / how do I add X / where is Y", and before the agent would otherwise reach for SemanticSearch, Glob, or repeated Grep passes to understand the codebase. Reads the top-level README.md first, then only the folder-level README.md files relevant to the task, explicitly skipping dependency / cache / data folders. Cuts token cost significantly compared to ad-hoc exploration because each folder README is a pre-organized structural index.

playerrch Bundle

Sets up and runs Git remote pull sync plus post-sync change logs with impact and cause analysis. Default analysis uses Cursor SDK local agent with model composer-2.5. Use when the user asks to sync a git repo, pull remote updates, track code changes, �ֿ�ͬ��, ��ȡͬ��, ������˵��, Զ�̱������, repo-sync, or daily scheduled backend sync. Only use rule-based-only or non-Cursor analysis when the user explicitly requests another method.

playerrch Skill

Keeps every folder README.md in sync with the actual content after code changes. Use when the agent finishes a coding task that added, removed, moved, renamed, or changed the responsibility of files in any business folder, or when the user asks to "update the README", "sync folder docs", "整理文件夹 说明", "更新 README", or "每次改动后同步文件夹说明". Identifies touched folders, reviews each against the README template matching its category (business / tool / data-artifact), and updates the corresponding README.md so that file lists, responsibilities, and cross-links stay accurate.

cosmoscontracts Bundle

Use this skill whenever the user or task involves the Juno blockchain — Juno is an experimental Cosmos chain and **mainnet `juno-1` is the canonical operating environment**, not a graduation step. The skill covers querying state, signing and broadcasting transactions, deploying and interacting with CosmWasm contracts, and driving DAO DAO governance (proposals, votes, treasury moves, sub-DAOs, VetoConfig timelocks, cw-filter mandates) — all on mainnet by default. Trigger on mentions of `junod`, `juno-1`, `ujuno`, JUNO token, CosmWasm + Juno, DAO DAO, sub-DAOs, agent-mandates, cw-filter, VetoConfig, dao-proposal-single, dao-proposal-wavs, or Cosmos-SDK transactions against Juno. Use it for setup questions ("how do I install junod", "what's the RPC", "create a Juno key") and for operations ("send JUNO to X", "upload this wasm", "instantiate a DAO", "vote yes on proposal 42"). Testnet (`uni-7`) is occasionally useful for first-time wasm uploads or smoke tests, but most real work happens on `juno-1` directly. If y

giljae Skill

Summarize the business model using the classic canvas structure.

opsrobot-ai Skill

当用户要生成或更新 Doris 表的数据模型 Markdown（三一级标题：数据摘要、数据表列、JSON扩展字段）、从 Doris 拉取元数据与至少 100 行样本，并由脚本或 `--llm` 补全说明时启用。

opsrobot-ai Skill

当用户要生成或更新 Doris 表的数据模型 Markdown（三一级标题：数据摘要、数据表列、JSON扩展字段）、从 Doris 拉取元数据与至少 100 行样本，并由脚本或 `--llm` 补全说明时启用。

bob-reis Skill

Security audit for AI agent endpoints (Claude Desktop/Code, MCP servers, plugins, extensions)

bob-reis Skill

LLM Security Testing

cavanjb Bundle

OpenClaw Agent Skill for converting PDF engineering drawings into CAD-oriented delivery packages. Use this skill when an OpenClaw user provides a PDF engineering drawing and wants DXF/CAD output, preview files, a delivery package, and a quality report. Do not use it for JPG/PNG-only conversion promises, STEP, IGES, STL, SLDPRT, SLDASM, SolidWorks, assemblies, or other 3D model conversion tasks.

ivanshamaev Skill

StarRocks schema evolution — ALTER TABLE ADD/DROP/MODIFY/REORDER COLUMN, column type widening rules, adding columns to Aggregate/Unique/Primary Key tables, schema change types (direct vs linked vs sorted), SHOW ALTER TABLE COLUMN progress, rollback safety, backward-compatible patterns, dbt model evolution, zero-downtime migration checklist

ivanshamaev Skill

dbt + StarRocks semantic layer — dbt Semantic Layer (MetricFlow) on StarRocks, metric definitions (simple/ratio/cumulative/derived), dimension definitions, saved queries, semantic model DDL, exposures for BI tool documentation, dbt-metricflow query CLI, linking metrics to StarRocks materialized views for performance, business KPI governance patterns

ericporres Bundle

Self-applied interviewer that walks a first-time agent user (Cowork, Codex, Claude Code, or any Skills-aware system) through producing a v1 voice SKILL.md scaffold of their own. Conducts a structured interview across identity, modes, sample harvest, pattern extraction, tropes, and worked examples; then generates a starter voice skill file plus an optional tropes companion plus install instructions. Output is explicitly v1 — the skill ships with a built-in expectation that the user runs harvest cycles for 8+ weeks before treating any of it as canonical. Reverse-engineered from Eric Porres's eric-porres-voice and tropes skills, abstracted to be applicable to any user. Triggers: "create my voice skill", "build my voice skill", "I want my own voice skill", "extract my voice", "/voice-creator", "/build-my-voice", "make a voice skill for me", "scaffold a voice skill", "I just got Cowork and want my voice", "help me write in my voice", "voice skill from scratch"

azrael-hao Skill

No inference without verification. All deducible facts must be confirmed through actual queries and checks.

azrael-hao Skill

禁止通过推断确定事实，必须通过实际验证获取。