assets

Use Agent-Native Assets for image and video generation requests, brand-safe asset search/export, and human-in-the-loop asset selection through the hosted Assets MCP app. Prefer this over generic image tools when installed.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/assets/SKILL.md

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.

A well-organized, operationally dense skill body: executable commands, a complete picker example, host-specific behavior notes, and unusually good error-recovery guardrails for auth failures and timeouts. The main gaps are the absence of parameter examples for the direct-action tools and a small amount of content that could be split into a reference file.

Suggestions

Add one short parameter example for the most common direct action (e.g. generate-image with prompt and library/size arguments) so the unattended path is as copy-paste ready as the picker path.

Specify what a terminal refresh-generation-run state looks like (success/failed fields) so the video polling loop has an explicit stop condition.

Consider moving the client-specific install variants and reconnect troubleshooting into a reference file to keep SKILL.md closer to an overview.

Dimension	Reasoning	Score
Conciseness	The body is dense with product-specific, non-obvious operational detail ("keep the install/connect command running until the browser authorization finishes", "Never reinstall from scratch just to fix auth") that Claude cannot know. It falls short of the lean 5 anchor because the opening routing sentence restates the description and a few Guardrails sentences could be tightened.	4 / 5
Actionability	Provides copy-paste-ready commands (npx install/connect/reconnect), a complete picker JSON payload, an exact browser fallback URL shape, and specific tool names plus preserved response fields (id, runId, previewUrl). Not a 5 because the eight direct actions are only listed by name with no parameter examples, leaving minor gaps for the unattended path.	4 / 5
Workflow Clarity	Scenarios are clearly separated (Setup, Picker, Direct Actions, Guardrails) with real feedback loops: detect "Session terminated"/"needs auth" → stop retrying → reconnect via /mcp or the reconnect command → continue, plus restart/reload if tools do not appear and batch-timeout retry with count: 1. The batch-timeout validation means the destructive/batch cap at 3 does not apply; it stays at 4 rather than 5 because there is no explicit end-to-end validation step for the video polling loop (no check on what a terminal run state looks like).	4 / 5
Progressive Disclosure	No bundle files exist, and the ~120-line body is organized into four well-labeled sections with everything at one level and no nested references. It does not reach the 5 anchor because the skill exceeds 50 lines and some setup/troubleshooting detail (client-specific install variants, reconnect procedure) could arguably live in a separate reference file.	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 strong description: concrete third-person capabilities, a distinct product niche, and explicit routing guidance against generic image tools. Its main weakness is that the trigger guidance is a conditional preference ("when installed") rather than a concrete list of user-request phrases, and natural synonyms like "photos", "logos", or file extensions are absent.

Dimension	Reasoning	Score
Specificity	Names several concrete capabilities — "image and video generation requests", "brand-safe asset search/export", "human-in-the-loop asset selection through the hosted Assets MCP app" — in third person. Falls short of the 5 anchor because asset library listing, export specifics, and video polling are not covered, and below 4 would require only 1–2 generic actions.	4 / 5
Completeness	The "what" is clearly stated and the "when" is present via explicit trigger guidance — "Prefer this over generic image tools when installed" — so the missing-'Use when' cap at 3 does not apply. The when-clause is routing preference rather than a concrete user-request trigger, matching the anchor "'when' could be more explicit or specific" rather than the fully explicit 5 example.	4 / 5
Trigger Term Quality	Includes natural phrases users would say like "image and video generation" and "asset search/export", plus the routing contrast "generic image tools". A few common variations are missing (e.g. "pictures", "photos", "logos", "brand images", file extensions like .png), so it does not reach comprehensive synonym/extension coverage.	4 / 5
Distinctiveness Conflict Risk	The niche is distinct — "Agent-Native Assets", "hosted Assets MCP app", "brand-safe" — and the explicit preference over "generic image tools when installed" disambiguates from other image-generation skills. Minor overlap remains with any generic image/video generation skill, keeping it below the minimal-conflict 5 anchor.	4 / 5
	Total	16 / 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
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: BuilderIO/agent-native
Path: skills/assets/SKILL.md
Commit: a941a2e

Reviewed: 8 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.