create-agent

Create a custom agent (.agent.md) for a specific job.

Quality

60%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./extensions/copilot/assets/prompts/skills/create-agent/SKILL.md

Quality

Content

67%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 a compact, well-structured instruction skill with a clear extract-clarify-iterate workflow and a genuine feedback loop, and it wastes almost no tokens. Its main weakness is actionability: the actual artifact creation is delegated to an external 'agents.md' template that is neither bundled nor summarized, leaving the most important step underspecified.

Suggestions

Inline a minimal .agent.md skeleton (frontmatter fields, one example section) so the drafting step is executable without hunting for 'agents.md'.

State where 'agents.md' lives (path or skill name) so the reference is resolvable from this skill alone.

Drop the closing reminder to follow 'agent-customization' guidelines since the opening already instructs loading them, removing the only redundancy.

Dimension	Reasoning	Score
Conciseness	The body is lean, directive, and free of concept over-explanation, but the 'agent-customization' pointer appears redundantly in both the opening ('Load and follow agents.md for template and principles') and the closing reminder, a minor instance that could be trimmed — the anchor for efficient with minor over-explanation.	4 / 5
Actionability	The extraction checklist and clarify-questions are concrete, but the core deliverable is left vague: 'Draft the agent file and save it' with no template, structure, or example inline, deferring entirely to an external 'agents.md' that is not part of this skill's bundle — matching the anchor for some concrete guidance but incomplete.	3 / 5
Workflow Clarity	The three-phase sequence (Extract from Conversation, Clarify if Needed, Iterate) is clear and includes a feedback checkpoint ('Identify the most ambiguous or weak parts and ask about those'), though final confirmation with the user before saving is only implicit — the anchor for a clear sequence with most checkpoints present.	4 / 5
Progressive Disclosure	Sections are well-organized with no inlined content that belongs elsewhere, and the 'agents.md' reference is clearly signaled and one level deep. However, the skill depends on that external file rather than being self-contained, so the simple-skill exception for a 5 (no need for external references) does not fully apply.	4 / 5
	Total	15 / 20 Passed

Description

53%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.

The description is concise and names a concrete, distinguishable artifact, but it states only a single action and gives no trigger guidance for when the skill should be invoked. It is serviceable but below the standard of the good examples, which pair a capability list with an explicit 'Use when...' clause.

Suggestions

Add a 'Use when...' clause, e.g. 'Use when the user wants to create, set up, or customize a subagent for a recurring job.'

List 1-2 more concrete actions (e.g. drafting the .agent.md, defining tool restrictions and persona) to raise specificity beyond a single verb.

Include natural synonyms users might say ('custom agent', 'subagent', 'specialized agent') to broaden trigger term coverage.

Dimension	Reasoning	Score
Specificity	Names one concrete action and artifact ('Create a custom agent (.agent.md)'), but offers only a single action with no elaboration of what the process involves, matching the anchor for domain plus 1-2 concrete actions rather than the several actions required for a 4.	3 / 5
Completeness	Has a clear 'what' (create a custom .agent.md) but no 'when' or 'Use when...' clause at all, which per the guidelines caps completeness at 3; not a 2 because the 'what' is concrete and unambiguous.	3 / 5
Trigger Term Quality	Contains relevant keywords ('custom agent', '.agent.md', 'specific job') but misses natural phrases and synonyms a user would actually say, such as 'build an agent', 'make a subagent', or 'customize my agent', fitting the anchor for some relevant keywords missing common variations.	3 / 5
Distinctiveness Conflict Risk	The specific '.agent.md' artifact establishes a distinct niche, but there is minor overlap risk with the closely related 'agent-customization' skill referenced in the body, matching the 'mostly distinct' anchor rather than the minimal-conflict 5.	4 / 5
	Total	13 / 20 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: posit-dev/positron
Path: extensions/copilot/assets/prompts/skills/create-agent/SKILL.md
Commit: b0258bc

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

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.