Skills
Reusable instruction sets scoped to a workspace that Agent blocks load at runtime
Skills are reusable instruction sets that define specialized behaviors for Agent blocks. Create a skill once and attach it to any agent in any workflow within the same workspace. Skills use the shared TradingGoose export JSON format, so you can export them from the Skill Editor and import them through the Skill List widget.
Skill Definition
Every skill is stored as a SkillDefinition with four user-facing fields:
| Field | Rules | Purpose |
|---|---|---|
| name | 1-64 chars, trimmed on save | Workspace-unique display name shown to people and agents |
| description | 1-1024 chars | Tells agents and teammates what the skill does |
| content | 1-50 000 chars | The full prompt: instructions, context, rules, and examples |
| id | Auto-generated | Immutable reference used by workflows, APIs, and the runtime loader |
Each skill also carries workspaceId, userId, createdAt, and an optional updatedAt timestamp.
Creating and Editing Skills
Open the Skill List widget and choose New skill.
Select the new skill and open it in a paired Skill Editor.
Enter a name up to 64 characters, a short description, and the full content, then choose Save. Validation rejects empty fields and duplicate names within the workspace.
To update an existing skill, select it in the Skill Editor, change any field, and save. All workflows referencing that skill will pick up the new content on their next run.
To delete a skill, use its delete action in the Skill List. Deletion requires admin or write permission. The list disables this action when only one skill remains.
Import and Export
Skills use the shared TradingGoose export JSON format. Export a skill from the Skill Editor and import it from the Skill List widget. Imported skills are normalized automatically, and duplicate names are renamed so you can move them between workspaces without manual cleanup.
Using Skills in Agent Blocks
The Agent block exposes a Skills input (skill-input type). At configuration time, select one or more skills from the workspace. At runtime, the agent first receives each available skill's name, description, and immutable ID. It can then load the full instructions for a relevant skill on demand.
Skills are workspace-scoped. Every team member with access to the workspace can browse, use, and (with write permission) edit or delete skills.
How Skills Work at Runtime
- The workflow executor reads the Agent block's selected skill references.
- TradingGoose resolves the selected skills in the execution's workspace and adds their names, descriptions, and IDs to the agent's context.
- When a skill is relevant, the agent calls its internal skill loader with the exact immutable ID.
- The loader retrieves that skill's current content and returns it as specialized instructions for the run.
Unavailable skill references are skipped. The display name helps the agent choose a skill, but the immutable ID is the runtime lookup key.
Permissions
| Action | Required Permission |
|---|---|
| List / read skills | Any workspace access |
| Create / update / delete | admin or write |
Best Practices
- One skill per role -- keep each skill focused on a single persona or task type.
- Include examples in the content to anchor the agent's behavior.
- Iterate in place -- editing a skill updates every workflow that references it, no re-wiring needed.
- Use descriptive names -- keep them short, trimmed, and unique within the workspace.