Skills that turn agents into dependable teammates.

OpenAI harmony response format

The [`gpt-oss` models](https://openai.com/open-models) were trained on the harmony response format for defining conversation structures, generating reasoning output and structuring function calls. If you are not using `gpt-oss` directly but through an API or a provider like Ollama, you will not have

User guide for gpt-oss-safeguard

ROOST and OpenAI have prepared a guide that explains how to write policy prompts that maximize [gpt-oss-safeguard's](https://github.com/openai/gpt-oss-safeguard) reasoning power, choose the right policy length for deep analysis, and integrate oss-safeguard's reasoning outputs into production Trust &

Techniques to improve reliability

When GPT-3 fails on a task, what should you do?

> Note: This repository contains Anthropic's implementation of skills for Claude. For information about the Agent Skills standard, see [agentskills.io](http://agentskills.io).

[![skills.sh](https://skills.sh/b/anthropics/skills)](https://skills.sh/anthropics/skills)

Anthropic Computer Use Demo

> [!NOTE]

Hermes Agent Skills gstack Browser Skills Taste Skill for Claude Skill Creator Pi Agent Superpowers Skills Impeccable Claude Skills Agent Browser Skills

Documentation

This directory contains the documentation source code for LlamaIndex, available at https://docs.llamaindex.ai.

Built for clarity

KISS content standard

Always expanding

Crawler + community

Popular collections

Skill formats

Four ways to package an agent skill

Browse the library

Prompt

Tight instructions that lock in tone, scope, and output shape.

Tool

Concrete actions, APIs, and schemas that your agent can call.

Workflow

Stepwise playbooks with decisions, checkpoints, and outputs.

Config

Settings and system prompts for reliable, repeatable behavior.

Popular skills

View all

PromptBeginner5 minmarkdownQuality: 72

Techniques to improve reliability

When GPT-3 fails on a task, what should you do?

PromptBeginner5 minmarkdownQuality: 72

User guide for gpt-oss-safeguard

PromptBeginner5 minmarkdownQuality: 72

OpenAI harmony response format

PromptBeginner5 minmarkdownQuality: 70

Anthropic Computer Use Demo

> [!NOTE]

PromptBeginner5 minmarkdownQuality: 70

> Note: This repository contains Anthropic's implementation of skills for Claude. For information about the Agent Skills standard, see [agentskills.io](http://agentskills.io).

[![skills.sh](https://skills.sh/b/anthropics/skills)](https://skills.sh/anthropics/skills)

Jan 15, 2026

PromptBeginner5 minmarkdownQuality: 69

Documentation

This directory contains the documentation source code for LlamaIndex, available at https://docs.llamaindex.ai.

Feb 1, 2026

Recently added

View all

Ultracite.mdc

Ultracite Rules - AI-Ready Formatter and Linter

请用中文总结下面这篇论文，输出严格 JSON，不要使用 Markdown。

JSON 字段：

Prompt Templates

LLM 摘要默认使用 `prompts/summary.zh.md`。可以在 `config.toml` 的 `[llm]`

Provider Setup Check

Validate and debug the AI provider configuration for this repo. Use when the app fails to start due to a missing API key, wrong provider name, connection error to an LLM endpoint, or when switching between providers (openai, openrouter, ollama).

Prompt Optimizer

Analyze, diagnose, and improve Pydantic AI system prompts and user prompts in this repo. Use when asked to improve output quality, fix inconsistent campaign generation, tune prompt wording, or reduce hallucinations and field truncation issues.

Pdf Debug

Debug WeasyPrint PDF generation issues in this repo. Use when asked to fix, investigate, or preview the PDF output, Jinja2 template rendering, or WeasyPrint errors such as missing fonts or native library errors (libpango, libcairo, libgdk-pixbuf).