Skip to content

Skills

Enable agents to produce downloadable document artifacts directly from chat — PDF, Spreadsheet, PowerPoint, and Word files generated mid-conversation that the user can preview and save locally.

Overview

A Skill is a per-agent capability that lets the agent generate a finished document artifact in response to a chat turn — for example, "Build me a Q3 sales report as a spreadsheet" or "Draft a one-page customer summary as a PDF". The artifact appears in the chat with preview + download affordances.

Skills are defined at the Global Level (platform-wide), made available to a workspace, and then enabled per agent by a workspace admin under Agent Settings > Skills.

Key Capabilities

Capability Detail
Supported artifact types PDF, PowerPoint, Spreadsheet, Word Document
Opt-in per agent Skills are off by default; each agent must explicitly enable the skills it needs. Click any skill chip in Agent Settings to toggle it on or off
Model-agnostic Works across Anthropic Claude, Google Gemini, and OpenAI ChatGPT model families
Token-budget aware Skill-enabled agents need a generous Max Output Tokens setting. Insufficient budget surfaces a clear error rather than a truncated artifact
RBAC-controlled configuration Permission to enable Skills on an agent is gated by RBAC. Enabled by default for the seeded Admin role

Configuration

Step 1: Skills available at the Global Level

Skills are added at the Global Level by the platform team. Once a Skill is added for a workspace, it becomes available for any agent in that workspace to enable.

Step 2: Enable Skills on an agent

  1. Open the agent under My Agents > (agent) > Settings
  2. Expand the Skills section
  3. Click any skill chip — PDF, PowerPoint, Spreadsheet, Word Document — to toggle it on or off. Enabled skills show a ✓ marker
  4. Click Save

Skills section under Agent Settings Agent Settings → Skills. The in-app banner reminds you: "For skills to work properly, ensure the configured models are with a sufficiently high output token limit." Each chip is a one-click toggle

Step 3: Verify Max Output Tokens

Because skill output is a complete document (not a short reply), the underlying language model needs enough output budget to produce the full artifact. Confirm under Agent Settings > Models > Add Language Model that Max Output Tokens is set generously — the platform default of 10,000 is a sane starting point for most artifact sizes.

Insufficient token budget

The Skills section itself displays an in-product reminder: "For skills to work properly, ensure the configured models are with a sufficiently high output token limit." If the configured Max Output Tokens cannot fit the artifact, the agent returns a clear error to the user rather than silently truncating the document. Raise the budget on the agent's language model configuration and retry.

RBAC

The ability to configure Skills on an agent is permission-gated. Skills is exposed as a sub-toggle under the Applicable for Update Settings group of an agent role — alongside other update-settings sub-permissions like Agent Info, Models, Tools, Behavior, Post Processors, etc.

A role needs both:

  1. Update Settings in the General permission group of the role, AND
  2. The Skills chip toggled on under Applicable for Update Settings
Status Roles
Enabled by default Seeded Admin role
Granted to other roles Toggle the Skills chip under Update Settings when editing the role

Skills RBAC permission Workspaces → Roles → Edit role. **Skills* appears as a chip under Applicable for Update Settings, between Tools and Post Processors*

Agent-type relevance

A note on the same page documents that not all permissions apply to every agent type. The Skills sub-toggle is most relevant for agent types that produce generated output (Chat Agent, Generative Agent). For RBAC fundamentals, see Workspaces > RBAC & Roles.

End-User Experience

Once a Skill is enabled on an agent, end users can leverage it from any chat — Web App, Sidebar, or any modality where the agent is published.

Typical flow:

  1. User asks the agent to produce a document (e.g., "Can you provide me a short write-up on products and services as a pptx?")
  2. The agent generates the artifact, summarises the contents in chat, and renders an artifact card below the response
  3. The artifact card shows the file-type icon, document title, filename + size (e.g., fifth-element-ai-platform-overview.pptx · 110.3 KB), and a prominent Download button
  4. The user clicks Download to save the file locally and open it in their desktop app

Skill artifact rendered in chat Agent chat — the agent summarises what it built, then renders the artifact card with filename, size, and a Download button. Show less collapses the response body if it's long

The artifact is bundled into the same response turn — no external service round-trip — so generation latency is bound by the LLM's output rate.

When to use Skills

Use case Recommended Skill
Sales/customer summaries, executive briefs PDF
Sales decks, board updates, training material PowerPoint
Data tables, pricing exports, multi-row reports Spreadsheet
Drafts, contracts, policy documents Word Document

Skills are best suited to structured, single-document outputs — not multi-file workflows or live editing surfaces.

Limitations

  • Skills are opt-in per agent — toggling a Skill on a parent template does not automatically enable it on cloned agents
  • Artifact size is bounded by the model's Max Output Tokens — very large documents may need to be generated in sections via follow-up turns
  • One artifact per turn — generating multiple document types in one response is not supported