Skill
A specialized capability module built into or extended on the ChatGPT platform, functioning similarly to a plugin mechanism. Skills augment the model with domain-specific capabilities such as image generation and document processing.
Core Facts
A Skill is a modular encapsulation that grants an agent specialized capabilities, containing complete workflows, judgment logic, and contextual knowledge, and can combine multiple tools to accomplish complex goals.
A Skill is a modular encapsulation that packages specialized capabilities, domain knowledge, and operational norms into reusable units, enabling Agents to invoke them automatically at the appropriate moment.
An AI Skill is a reusable AI execution unit that encapsulates the logic of a specific task, essentially a combination of structured prompts and code logic
Timeline (last 90 days)
In security scenarios, Skills can solidify detection approaches, payload construction methods, and report templates for common vulnerabilities into reusable modules, enabling agents to invoke a unified standard process.
A Skill's on-demand loading mechanism is consistent with the retrieval-augmented generation (RAG) approach: retrieve first, then inject, then execute.
A promotional video production Skill went through more than a dozen iterations before the final version that could stably produce good videos was open-sourced.
One of Li Dengdeng's Skills helped someone else earn 1,000 yuan; that Skill teaches an Agent to generate high-quality promotional videos with code.
A large number of verified, ready-to-run Skills can be found on major domestic social media, covering scenarios such as making images, PPTs, posters, and videos.
When the same Skill is given to different Agents to execute the same task, the results differ markedly: some Agents skip most of the steps and deliver quickly, while others follow every step but differ greatly in layout and aesthetics.
Evaluation Driven Development is an advanced technique: before building a Skill, first have the AI attempt the task bare, without giving it any methodology, and observe where it gets stuck—those sticking points are the real gaps that should be filled in by hand.
LLMs exhibit Context Rot: the more tokens loaded, the lower the recall precision; the official recommendation is that a Skill exceeding 500 lines should be split, while practitioners tend to keep it within 200 lines.
The root cause of unstable output quality is often writing the SOP too narrowly; the correct approach is to give principles rather than lock down steps—write strict steps for stages with objective right-or-wrong standards, and write principle frameworks for stages that depend on judgment.
A Skill's Name and Description are the only basis on which the Agent judges whether to trigger the Skill; names should use lowercase letters and hyphens linking verbs or nouns in series (such as weekly-report-drafting), and the Description must contain both "what it does" and "when to use it", use the third person, and use trigger words that are what users would naturally say.
40 more timeline events
All Facts (20)
A Skill is a modular encapsulation that grants an agent specialized capabilities, containing complete workflows, judgment logic, and contextual knowledge, and can combine multiple tools to accomplish complex goals.
90%VerifiedA Skill is a modular encapsulation that packages specialized capabilities, domain knowledge, and operational norms into reusable units, enabling Agents to invoke them automatically at the appropriate moment.
90%VerifiedAn AI Skill is a reusable AI execution unit that encapsulates the logic of a specific task, essentially a combination of structured prompts and code logic
90%VerifiedA Skill is the concept of packaging multi-step interaction logic, context management, and output constraints into a callable unit, distinct from a single-interaction prompt template.
80%VerifiedA complete Skill contains a Skill.md main document, a Reference folder, a Script folder, and an Assets folder.
80%VerifiedAfter completing a task, Hermes Agent automatically creates new Skills and updates user information.
80%VerifiedAI Agent基于ReAct（Reasoning + Acting）等框架，能够将复杂任务分解为多个子步骤，自主决定何时调用外部工具
80%VerifiedSkill与普通提示词的区别在于Skill是可复用、可积累的结构化能力单元
80%VerifiedSkills本质上是结构化的系统提示词（System Prompt）集合，通过标准化接口注入到AI代理的上下文中
80%VerifiedSkills are not competitors to MCP but rather a layer above MCP: MCP is responsible for connecting tools and standardizing external service interfaces, while Skills are responsible for packaging and encapsulating complete task execution capabilities.
80%VerifiedAgent Skills是Anthropic发布的能力封装标准，定义为一个包含Skill.md文件的目录，可包含说明、脚本和其他资源
80%VerifiedAgents (intelligent agents) can invoke tools on a computer and independently complete an entire set of complex tasks, unlike chatbots that can only conduct one-round-at-a-time conversations.
75%VerifiedSkills are essentially Markdown files stored in the project's .claude/ directory.
75%VerifiedWorkBuddy has a built-in Chinese Skill marketplace where common capabilities can be installed with one click.
70%VerifiedAccording to the presenter's observations, programmers at large tech companies may spend 70%–80% of their AI usage time writing Skills, with relatively little time spent directly writing code.
70%VerifiedAgent Skill adopts a lazy-loading mechanism whereby the Agent first reads only the metadata (name and description) of each Skill and loads the full instructions only after matching the user's need, so having multiple Skills coexisting does not increase Token consumption
70%VerifiedA Skill is a folder containing a file named skill.md, whose description resides in the agent context at low cost, with the body loaded only when a task matches.
65%VerifiedWhen a skill description is so detailed that it contains all the information needed to complete a task, the model will skip the skill-calling step.
65%VerifiedThe Skill identification process typically relies on the model's semantic understanding of capability descriptions, and how a Skill description is written has a greater impact on actual invocation outcomes than the complexity of the code implementation.
65%VerifiedThe Agent + Skill combination is a "brain + arms" architecture: the Agent is responsible for understanding intent and formulating plans, while the Skill is responsible for concrete execution.
65%