CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-creator

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.

95

1.28x
Quality

92%

Does it follow best practices?

Impact

100%

1.28x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Evaluation results

100%

12%

Automated Security Review Agent Setup

Subagent context memory and invocation for analysis agents

Criteria
Without context
With context

Subagent enabled

100%

100%

Full context memory

33%

100%

Automatic invocation

100%

100%

Red color

100%

100%

Specialized system prompt

100%

100%

Bash never

100%

100%

File edit never

100%

100%

Three-dash key format

100%

100%

Valid tool keys only

100%

100%

100%

33%

Documentation Generator Agent for a Growing API Platform

Tool approval key format and approval strategy

Criteria
Without context
With context

Subagent enabled

100%

100%

Last-message context memory

0%

100%

Automatic invocation

100%

100%

Purple/creative color

0%

100%

Bash never

100%

100%

Three-dash separator

100%

100%

Only recognized tool keys

76%

100%

Read/write tools configured

100%

100%

100%

19%

Local LLM Developer Agent for an Isolated Project

Provider/model split and profile storage location

Criteria
Without context
With context

Provider is ollama

100%

100%

Model is llama3.2

100%

100%

Project-specific path

50%

100%

Off context memory

100%

100%

On-demand invocation

100%

100%

No temperature set

100%

100%

No maxTokens set

100%

100%

Subagent enabled

0%

100%

Repository
hotovo/aider-desk
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.