Tessl • The management layer for AI agent skills

SkillPublic

create-agent

Comprehensive guide for creating Claude Code agents with proper structure, triggering conditions, system prompts, and validation - combines official Anthropic best practices with proven patterns

Quality

62%

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 ./antigravity/skills/create-agent/SKILL.md

The canonical home for this skill is create-agent in NeoLabHQ/context-engineering-kit

Quality

Content

56%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 complete copy-paste agent templates and a clear, validated creation workflow, but it suffers from heavy redundancy across repeated templates and summaries and makes no use of progressive disclosure — everything lives in one ~660-line file with a dangling script reference. Trimming repetition and splitting examples into reference files would meaningfully improve both conciseness and navigation.

Suggestions

Consolidate the repeated frontmatter/template explanations into a single canonical section and reference it elsewhere to eliminate roughly four near-duplicate renderings.

Move the two full production example agents into a references/ file (e.g. references/examples.md) and link to it, reducing the main body and adding real progressive disclosure.

Either provide the referenced scripts/validate-agent.sh in a scripts/ bundle or remove/descope the validation-script invocations so references resolve to real files.

Dimension	Reasoning	Score
Conciseness	The ~660-line body is noticeably verbose: the frontmatter template is repeated roughly four times (Agent File Structure, Frontmatter Fields Reference, two Production Examples, Agent Creation Process), two full example agents are inlined, and DO/DON'T summaries restate earlier sections — several padded/redundant sections beyond 'some unnecessary explanation'.	2 / 5
Actionability	Provides copy-paste-ready full agent files, concrete bash commands (mkdir, touch), a validation invocation, and valid/invalid tables, giving mostly executable guidance with only minor gaps (the referenced validate-agent.sh content is never shown).	4 / 5
Workflow Clarity	A clear 6-step Agent Creation Process sequences gather → create → write frontmatter → write prompt → validate → test-triggering, with explicit validation and testing checkpoints and checklists, but the fix-and-revalidate feedback loop is only implicit rather than spelled out.	4 / 5
Progressive Disclosure	Section headers are well-organized, but the skill is monolithic with no bundle files (references/, scripts/, assets/ absent) and content that could be split (full example agents, detailed frontmatter reference) is inlined; the one external reference (scripts/validate-agent.sh) points to a non-existent file.	3 / 5
	Total	13 / 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 clearly conveys what the skill does and occupies a distinct niche, but it reads as a feature list padded with buzzwords ('Comprehensive guide', 'official Anthropic best practices') and critically omits any 'when to use' trigger guidance. Adding a 'Use when...' clause with natural trigger phrases would substantially raise completeness and trigger quality.

Suggestions

Add an explicit trigger clause such as 'Use when creating, refining, or validating Claude Code agents (agents/*.md files)' to satisfy the completeness 'when' requirement.

Replace buzzword padding ('Comprehensive guide', 'combines official Anthropic best practices with proven patterns') with concrete actions, e.g. 'Scaffold agent frontmatter, write system prompts, and run structure validation'.

Include natural trigger synonyms users actually say (e.g. 'create an agent', 'agent definition', 'agents/*.md') to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ('creating Claude Code agents') and lists several concrete elements ('structure, triggering conditions, system prompts, and validation'), but 'Comprehensive guide' is generic framing and the listed items are attributes rather than distinct actions, landing between the minimal-action and 1-2-concrete-actions anchors.	3 / 5
Completeness	The 'what' is clear (a guide for creating agents with named components) but the 'when' is entirely absent with no 'Use when...' clause or equivalent, so completeness is capped at 3 per the missing-trigger-guidance rule.	3 / 5
Trigger Term Quality	'creating Claude Code agents' is a natural phrase a user might say, but there are no synonyms, variations, or file extensions and no explicit trigger clause, matching 'some relevant keywords but missing common variations'.	3 / 5
Distinctiveness Conflict Risk	Agent creation is a clear, distinct niche unlikely to fire for unrelated skills, though the lack of explicit triggers leaves minor overlap risk with adjacent plugin-dev skills.	4 / 5
	Total	13 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (668 lines); consider splitting into references/ and linking	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	14 / 16 Passed

Repository: NeoLabHQ/context-engineering-kit
Path: antigravity/skills/create-agent/SKILL.md
Revision: 23e2428

Reviewed: 26 days ago

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.

create-agent

Table of Contents

Is this your skill?