// Create and configure AiderDesk agent profiles by defining tool groups, approval rules, subagent settings, and provider/model selection. Use when setting up a new agent, creating a profile, or configuring agent tools and permissions.
Create AiderDesk extensions by setting up extension files, defining metadata, implementing Extension interface methods, and updating documentation. Use when building a new extension, creating extension commands, tools, or event handlers.
Create new AiderDesk UI themes by defining SCSS color variables, registering theme types, and adding i18n display names. Use when adding a theme, creating a color scheme, customizing appearance, or implementing dark mode and light mode variants.
Write unit tests, component tests, and integration tests for AiderDesk using Vitest and React Testing Library. Use when creating new tests, adding test coverage, configuring mocks, setting up test files, or debugging failing tests.
| name | agent-creator |
| description | Create and configure AiderDesk agent profiles by defining tool groups, approval rules, subagent settings, and provider/model selection. Use when setting up a new agent, creating a profile, or configuring agent tools and permissions. |
Create agent profiles stored as config.json in ~/.aider-desk/agents/{name}/ (global) or {project}/.aider-desk/agents/{name}/ (project-level).
Read all reference files before proposing a profile to ensure accuracy.
Ask user: "Describe the agent's purpose and what it should do."
Based on their response, internally propose all properties:
Ask: "Which provider/model? (format: provider/model, e.g., anthropic/claude-sonnet-4-5-20250929)"
ONLY ask if user specifically mentioned temperature, max tokens, rate limiting, or MCP servers:
If not mentioned, skip this entire step.
Ask: "Global profile (all projects) or project-specific? (default: global)"
Present one complete summary with all proposed properties. Ask:
"Here's your agent profile. Should I create it?"
Show:
DO NOT ask for confirmations on individual items. Only one final approval.
On user confirmation, generate the profile. Verify structure against references/profile-examples.md before creating files.
Keys: {group}---{tool} (three dashes)
Default is "ask". Only set:
power---bash for read-only agents)Only include tools that exist in references/tool-approval-guide.md.
Every agent is a subagent (enabled: true). See references/subagent-guide.md for detailed guidance.
contextMemory: Default is "off" (fresh each time)
"full-context" only for specialized analysis agents (code review, security audit)"last-message" for progressive builders (documentation, iterative tasks)systemPrompt: Specialized for agent's purposeinvocationMode: automatic (specialized), on-demand (general)color: Relevant color (e.g., red=security, blue=power tools)description: Clear description for auto-invocation{
"name": "my-agent",
"provider": "anthropic",
"model": "claude-sonnet-4-5-20250929",
"maxIterations": 250,
"toolGroups": ["power", "todo"],
"toolApprovals": {
"power---bash": "ask",
"power---read": "always"
},
"subagent": {
"enabled": true,
"contextMemory": "off"
}
}
{group}---{tool} formatreferences/agent-profile-schema.md - Complete schemareferences/subagent-guide.md - Subagent configuration guide (context memory modes)references/profile-examples.md - Examplesreferences/tool-approval-guide.md - Tool approval configassets/templates/config.json.template - Templateassets/examples/sample-profile.json - Example