CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

create-subagent

Create custom subagents for specialized AI tasks. Use when you want to create a new type of subagent, set up task-specific agents, configure code reviewers, debuggers, or domain-specific assistants with custom prompts.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.cursor/skills-cursor/create-subagent/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with executable commands and complete copy-paste examples, organized into a clear sequenced workflow. Its weaknesses are redundancy across the three full examples and repeated best-practices guidance, plus a thin validation step.

Suggestions

Cut to one or two example subagents — the data-scientist example largely duplicates what the code-reviewer and debugger examples already demonstrate, saving ~30 lines.

Merge the 'Best Practices' section into 'Writing Effective Descriptions' and the workflow steps to eliminate repeated guidance about descriptions and proactive language.

Strengthen the validation checkpoint with a concrete check, e.g. re-reading the created file to confirm the frontmatter parses (or a YAML lint command) before testing the agent.

DimensionReasoningScore

Conciseness

Mostly efficient, but the three full-length example subagents (~90 lines) overlap heavily — the data-scientist example adds little beyond the first two — and the 'Best Practices' section repeats guidance already given in 'Writing Effective Descriptions' ('Write detailed descriptions', 'Use proactive language').

3 / 5

Actionability

Fully executable guidance throughout: copy-paste-ready example agent files, concrete bash commands ('mkdir -p .cursor/agents', 'touch .cursor/agents/my-agent.md'), a required-fields table, and a specific test invocation ('Use the my-agent subagent to [task description]') covering the common cases.

5 / 5

Workflow Clarity

The 5-step workflow (decide scope, create file, define configuration, write system prompt, test) is clearly sequenced with a test step and a troubleshooting recovery path; however validation is light — 'Verify YAML frontmatter syntax is valid' names no concrete check or command.

4 / 5

Progressive Disclosure

With no bundle files present, the single-file structure is well organized (clear headers, a locations table, distinct sections) and easy to navigate; the only content arguably belonging in a separate file is the three inlined full example agents.

4 / 5

Total

16

/

20

Passed

Description

75%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A solid description with an explicit what/when structure and natural trigger phrasing for the subagent-creation niche. Its main weaknesses are the when-clause being nearly circular with the what, and slightly incomplete synonym coverage.

Suggestions

Rewrite the 'Use when' clause with situational triggers independent of the capability statement, e.g. 'Use when the user asks to create, customize, or automate a specialized agent, or mentions .cursor/agents/ or agent configuration files.'

Add common synonyms such as 'custom agent', 'specialized assistant', or '.cursor/agents' paths to broaden natural trigger coverage.

Briefly mention managing existing subagents (editing, sharing, or removing them) to round out capability coverage.

DimensionReasoningScore

Specificity

Lists several concrete actions ('create', 'set up', 'configure') targeting named agent types ('code reviewers, debuggers, or domain-specific assistants with custom prompts'), but coverage has minor gaps such as managing or editing existing subagents.

4 / 5

Completeness

Explicitly answers both what ('Create custom subagents for specialized AI tasks') and when ('Use when you want to create a new type of subagent...'), but the when-clause largely restates the what rather than adding independent situational triggers.

4 / 5

Trigger Term Quality

Natural phrases users would say appear ('create a new type of subagent', 'set up task-specific agents', 'code reviewers', 'debuggers'), but a few common synonyms such as 'custom agent' or '.cursor/agents' are missing.

4 / 5

Distinctiveness Conflict Risk

The niche of authoring subagent definition files is clearly distinct; the only minor overlap risk comes from naming generic roles like 'code reviewers' and 'debuggers', which could collide with skills that perform those tasks.

4 / 5

Total

16

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

saski/arnesto

Path

.cursor/skills-cursor/create-subagent/SKILL.md

Reviewed

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.