---
title: "Agent Skills: Reusable Playbooks Your Agents Load On Demand"
slug: "agent-skills"
date: "2026-08-11"
tag: "Application"
summary: "Agent Skills ship as a first-class surface: an organization-wide skill catalog, per-agent assignment, private skills, attachable file resources, and a / picker in chat that lets users invoke a skill by name."
author: "ibl.ai Engineering"
repo: "iblai/vibe"
linkedin: |
  Most "give your AI more capabilities" advice ends the same way: a system prompt that grows until every conversation carries instructions for jobs it will never do.

  Agent Skills take the other path. A skill is a reusable playbook — written instructions plus optional reference files — for exactly one job: research a topic, review code, draft a course spec. The agent loads it only when it's relevant, so one agent can carry dozens of capabilities without any single conversation paying for all of them.

  What shipped:

  → An organization-wide skill catalog. Write a skill once, attach it to as many agents as you want.
  → Private skills, scoped to a single agent, which take precedence over org skills with the same slug.
  → File resources per skill — references, scripts, and uploaded assets the agent can use.
  → A / picker in chat. Type "/" and pick the skill by name, the way you'd invoke a command.
  → The same surface in the SDK, so anyone building on ibl.ai gets it as a component rather than rebuilding it.

  And the part that matters underneath: on ibl.ai you own all the code and the data. The skills your team writes are instructions sitting in your own database, on your own infrastructure — not prompt assets locked inside someone else's product that you'd have to rewrite the day you switch vendors.

  #iblai #AgenticAI #EnterpriseAI #AIAgents
---

Agent Skills are now a first-class surface on the platform. A skill is a reusable playbook — written instructions plus optional reference files — that an agent follows for one specific job, such as researching a topic on the web, reviewing code, or drafting a course spec.

The point is selective loading. The agent reads a skill only when it is relevant to the task at hand, so a single agent can carry many capabilities without every conversation paying the token cost of all of them.

## The Skills panel

![The Skills panel on the Agent Skills tab, listing Image Creation v1.0.0 and Web Research v1.0.0 with enable toggles and a New Skill button](/images/updates/agent-skills-agent-tab.webp)

The panel splits into two tabs. **Agent Skills** is the agent's effective set — the skills assigned to it plus its own private skills, deduped by slug — each with an enable toggle and a row menu.

A note in the header states the rule that governs rollout: skills added or removed apply to new chat sessions only, while edits to a skill's instructions apply immediately, including in conversations already in progress. That split is deliberate. Changing what an agent *is* mid-conversation is disorienting; correcting *how* it does a job should not require anyone to start over.

## An organization-wide catalog

![The Available Skills tab showing the organization catalog, with green Added chips on skills already attached to this agent](/images/updates/agent-skills-available-tab.webp)

**Available Skills** is the organization's full catalog. Write a skill once and attach it to as many agents as you want with **Add to Agent**; skills already covered by the agent's set show a green **Added** chip instead.

![The row menu open with Remove from Agent, Edit, and Delete actions](/images/updates/agent-skills-row-actions.webp)

Each row carries the actions that match its scope. **Remove from Agent** detaches a skill from one agent and leaves it in the catalog for everyone else. **Edit** and **Delete** change the skill itself, organization-wide — and appear only for skills your organization owns, so featured skills stay read-only rather than failing at save time.

## Private skills and slug precedence

![The New Skill dialog with an Only This Agent toggle and fields for Name, Slug, Version, Category, Description, and Instruction](/images/updates/agent-skills-new-skill.webp)

A skill created with **Only This Agent** is private to that agent, and it takes precedence over an organization skill with the same slug. That gives teams a clean override path: publish a standard `web-research` playbook for the whole organization, and let one agent run a specialized version of it under the same name, with no forking and no rename.

![The Edit Skill dialog on the General tab, showing the saved skill name and slug](/images/updates/agent-skills-edit-general.webp)

Skills carry a name, slug, version, category, description, and the instruction text itself — the playbook the agent follows.

## File resources

![The Edit Skill dialog on the Resources tab, with a New Resource button and a course_spec.example.yaml file tagged Reference](/images/updates/agent-skills-resources.webp)

Instructions are often not enough on their own. A skill can carry files: **reference** and **script** resources are text, and **assets** are uploaded files. A course-drafting skill can ship the YAML spec it targets; a code-review skill can ship the style guide it enforces.

## Invoking a skill from chat

![The chat composer with a slash typed, showing a picker listing Image Creation /image-creation and Web Research /web-research](/images/updates/agent-skills-slash-picker.webp)

Skills are discoverable in the conversation itself. Typing `/` as the first word of a message opens a picker listing the agent's enabled skills by name and slug; arrow keys browse, Enter or Tab selects, and Esc dismisses it. Plain text that happens to start with `/` is never blocked.

![The chat page with the skill picker open above the composer](/images/updates/agent-skills-chat.webp)

## Available in the SDK

The same surface ships to developers building on the platform. `AgentSkills` is an independent component — it takes a platform key and an agent ID and manages its own fetching, dialogs, and pagination — and the chat `/` picker turns on with a single prop, or accepts a host-supplied skill list for custom composers. The picker's pieces are exported individually for anyone building their own composer.

Agent Skills apply to Base Agent agents, and are managed independently of the Claw sandbox — no sandbox instance is required.

Documentation: [Agent Settings: Skills](https://ibl.ai/docs/os/agent-settings/skills).