惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

人人都是产品经理
人人都是产品经理
博客园_首页
博客园 - 三生石上(FineUI控件)
V
Visual Studio Blog
Hugging Face - Blog
Hugging Face - Blog
美团技术团队
小众软件
小众软件
T
Tailwind CSS Blog
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
月光博客
月光博客
有赞技术团队
有赞技术团队
WordPress大学
WordPress大学
博客园 - 【当耐特】
Apple Machine Learning Research
Apple Machine Learning Research
罗磊的独立博客
V
V2EX
酷 壳 – CoolShell
酷 壳 – CoolShell
IT之家
IT之家
量子位
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Recent Announcements
Recent Announcements
M
MIT News - Artificial intelligence
阮一峰的网络日志
阮一峰的网络日志
The GitHub Blog
The GitHub Blog

Mozilla.ai

Open Source Is Not a Virtue: It’s an Ownership Model Using Octonous as a Product Operations Manager llamafile v0.10.5 The Future of AI is Personal — But Not The Way You Think How Frontier Labs Are Building Subtle Developer Lock-In Who Cares About LLM costs? Stop Chasing New Models. Build Once and Access Them All. From Evaluation to Guardrails: What We Brought to ACM FAccT 2026 Open Models are ready for agents. Their APIs are not. Using Octonous as an AI Safety Engineer The Control Layer: Why the Next Era of AI Is About Infrastructure, Not Just Models Introducing Otari: The Open-Source LLM Control Plane Announcing transcribe.cpp Using Octonous as a Product Manager Image Classification Comes to encoderfile What is an LLM control plane? Use the Otari Gateway with OpenCode Otari: Own Your AI Stack | AI Gateway & Hosted Platform AI Got Expensive. Now What? | Mozilla.ai cq exchange: Agents without Borders The Interface Is No Longer the Product VIBE✓: First Defense for cq (Stack Overflow for Agents) Octonous Open Beta: What We've Learned and Where We're Going Sovereign AI: Control, Choice, and Beyond Geopolitics Encoderfile’s New Format: Why a “Dull” Design Wins The Real Challenge Behind Small Trade Businesses Hardening Your LLM Dependency Supply Chain cq: Stack Overflow for Agents cq: Stack Overflow for Agents llamafile Reloaded: What’s New in v0.10.0
Introducing Agent Skills in Octonous
Hareesh Bahuleyan · 2026-08-25 · via Mozilla.ai

Agent Skills in Octonous offer a central, reusable library for your context, guidelines, and procedures. Written in standard Markdown, Skills end repetitive setup. Store team knowledge once, share it across your workspace, and seamlessly attach it to automated tasks or live conversations.

Hareesh Bahuleyan

2 min read

Introducing Agent Skills in Octonous

Agent Skills provide a way to store reusable instructions and context in Markdown files. Octonous now supports Skills, enabling you to maintain a personal or team-wide shared Skills library that can be invoked when relevant to a task or automation.

Skills can be used for writing guidelines, product information, research methods, operational procedures, review criteria, and more. Keeping such information in a Skill reduces repeated setup and provides a shared reference for related tasks.

Skills

Prompts are useful for task-specific instructions, but they are not always the most practical place to maintain information that applies across many tasks. Details such as terminology, workflow steps, quality criteria, and organizational context may need to be reused by several people or workflows.

A Skill is a named set of instructions or context that agents can load when it is relevant to a task. Storing it separately can make it easier to review, update, and apply consistently. Once created, it can be reused across conversations and workflows. Skills are written in plain Markdown and follow the open Agent Skills format

Creating or Importing a Skill

Octonous supports 3 ways to add Skills to your library:

  1. Write one manually. Create a new Skill by giving it a name and description, along with the necessary instructions.
  2. Draft it with Octonous. Octonous includes a built-in Skill called `skill-builder`. It asks relevant clarifying questions about the workflow, expected outcomes, and constraints, then produces a Skill draft for you to review and approve.
  3. Import it. Alternatively, users can provide the GitHub URL of a `SKILL.md` file or upload it directly. For example, here is a publicly available product marketing Skill:

0:00

/0:29

Workspace Sharing

New Skills are private by default, but they can also be shared with a workspace when they are intended for broader use.

A workspace-shared Skill gives team members a common source of instructions and context. When that information changes, the shared Skill can be updated in one place rather than requiring each user to maintain a separate copy.

Skills and Automations

Skills and automations serve related but distinct purposes.

Automations determine when a task runs, while Skills provide reusable instructions and context for how it runs.

For example, an automation can run a task on a schedule, such as drafting a newsletter every Thursday or summarizing campaign performance each month. Its configuration determines when the task runs.

Reusable context, such as product positioning, may apply to several automations as well as live conversations. Keeping it in a Skill allows the same context to be used across multiple workflows.

In Octonous, Skills can be attached to automations. Each scheduled run can then use the instructions and context from its attached Skills.

Try Skills in Octonous

Skills are now available to all Octonous users. Open the Skills page to create or import Skills for your agentic workflows, then invoke them in a conversation using the / slash command. Sign up to get started.