extensions

Creating, editing, and managing extensions — sandboxed Alpine.js mini-apps that run inside iframes. Use only when the host app explicitly enables extensions and the user asks for a one-off custom block or mini-app. Distinct from LLM "tools" (function calls) — see note below.

Quality

83%

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

Quality

Content

78%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.

A highly actionable, well-structured body with strong progressive disclosure and concrete code throughout. The main weakness is redundancy — the definition of what extensions are appears in two sections — which hurts token efficiency.

Suggestions

Merge the duplicate 'What extensions are' sections (CRITICAL block and the later '## What extensions are' heading) into a single canonical definition to remove redundancy.

Tighten the multi-paragraph terminology note into a compact table or a few bullets; the back-compat detail already lives in references/api.md.

Add an explicit validate-then-proceed checkpoint (e.g., verify the edit applied via get-extension or a render check) to the editing workflow to elevate workflow clarity.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and domain-specific, but the 'What extensions are' concept is defined twice (the CRITICAL section and again at line ~197) and the terminology note is lengthy; these could be tightened or merged.	3 / 5
Actionability	Provides copy-paste-ready bash, HTTP, and JSON payloads plus complete JavaScript/Alpine.js examples (extensionData usage, customerAnalyzer component) covering the common create/edit/persist cases.	5 / 5
Workflow Clarity	Sequences the create and edit flows with explicit safety acknowledgements (operation: "replace", expectedMatches, format: true) and rollback via history, but validation checkpoints are spread rather than uniformly explicit.	4 / 5
Progressive Disclosure	Keeps the overview inline and delegates exhaustive helper tables and full worked examples to real one-level-deep files (references/api.md, references/examples.md) that are clearly signaled in a References section with read-when guidance.	5 / 5
	Total	17 / 20 Passed

Description

87%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 strong description that explicitly states both capabilities and trigger conditions in third person, and actively disambiguates the extension primitive from LLM tools. Trigger-term coverage is good but could add a few more natural synonyms.

Dimension	Reasoning	Score
Specificity	Names the domain ('sandboxed Alpine.js mini-apps that run inside iframes') and lists several concrete actions ('Creating, editing, and managing'), though 'managing' is somewhat generic, leaving minor coverage gaps.	4 / 5
Completeness	Clearly answers what ('Creating, editing, and managing extensions — sandboxed Alpine.js mini-apps') and when ('Use only when the host app explicitly enables extensions and the user asks for a one-off custom block or mini-app') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural terms a user would say ('extension', 'custom block', 'mini-app', 'one-off') but misses some common synonyms/variations; good but not comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche and explicitly disambiguates from the closest confusable primitive ('Distinct from LLM "tools" (function calls)'), giving minimal conflict risk.	5 / 5
	Total	18 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (534 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	13 / 16 Passed

Repository: BuilderIO/agent-native
Path: .agents/skills/extensions/SKILL.md
Commit: a06e0ac

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.