stitch-loop

Iterative design-to-code feedback loop. Critique → adjust → ship cycle for tightening visual fidelity between brief and built UI.

Quality

38%

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 ./skills/stitch-loop/SKILL.md

Quality

Content

27%

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

The SKILL.md body is a thin catalogue stub: it advertises an upstream skill rather than providing the critique → adjust → ship workflow it names. It is organized and not padded, but it offers no executable guidance and no sequenced loop with validation checkpoints.

Suggestions

Inline the actual critique → adjust → ship workflow as numbered steps (e.g. 1. render the built UI, 2. compare against the design brief and list visual deltas, 3. revise the component, 4. re-render and verify the delta is resolved), with an explicit verify checkpoint.

Provide a concrete, executable example of one critique→adjust→ship cycle (a sample delta, the code change, and the re-check) instead of only linking to the upstream repo.

Trim the meta explanation about catalogue advertising and replace it with the skill's operating instructions, keeping the upstream link as a short 'Source' line.

Dimension	Reasoning	Score
Conciseness	The body is short and not padded with concepts Claude already knows, but it spends several lines explaining catalogue/advertising mechanics ('This catalogue entry advertises the skill in Open Design so the agent discovers it during planning') that are meta-context rather than skill-advancing guidance; it is mostly efficient but includes some text that could be tightened.	2 / 3
Actionability	The only executable instruction is the shell snippet `open https://github.com/...`, which merely points to an external URL rather than providing the actual critique → adjust → ship workflow; it describes the skill rather than instructing Claude how to perform it, matching the score-1 'describes rather than instructs' anchor.	1 / 3
Workflow Clarity	The promised 'Critique → adjust → ship' loop is never broken into concrete sequenced steps, and there is no validation/feedback checkpoint despite the body advertising an 'iterative feedback loop'; the steps are effectively missing, matching the score-1 anchor.	1 / 3
Progressive Disclosure	The body is organized into clear sections (What it does, Source, How to use) and points one level deep to an upstream URL, which is appropriate given no bundle files exist; however the only reference is an external repository link with no local reference files, so structure is present but navigation is shallow and points off-repo rather than to a bundled reference.	2 / 3
	Total	6 / 12 Passed

Description

50%

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 names a coherent niche (design-to-code fidelity loop) but keeps its actions abstract and relies on a separate triggers list rather than an explicit 'Use when...' clause. It is competent but not maximally concrete or trigger-explicit.

Suggestions

Add an explicit 'Use when...' clause to the description itself naming natural user phrases (e.g. 'Use when iterating between a design brief and built UI to tighten visual fidelity').

Replace abstract verbs ('critique', 'adjust') with concrete actions (e.g. 'compare the built UI against the design brief, identify visual deltas, revise the component, and re-verify').

Replace the jargon-heavy trigger 'fidelity loop' and broaden 'design iteration' with more common phrasings a user would naturally say, e.g. 'match UI to design', 'tighten visual fidelity'.

Dimension	Reasoning	Score
Specificity	The description names a domain ('design-to-code feedback loop') and the actions 'Critique → adjust → ship', but 'critique' and 'adjust' are abstract verbs that are not as concrete as the score-3 example's 'extract text, fill forms, merge documents'; it lands in the 'names domain and some actions, but not comprehensive' band rather than listing multiple concrete, specific actions.	2 / 3
Completeness	The description answers 'what it does' clearly, but the 'when to use it' trigger guidance lives only in a separate frontmatter `triggers` list rather than an explicit 'Use when...' clause in the description itself; per the judging guidelines a missing explicit trigger clause caps completeness at 2.	2 / 3
Trigger Term Quality	The frontmatter lists several relevant trigger phrases ('stitch loop', 'design to code', 'design iteration', 'fidelity loop'), giving some natural keyword coverage, but the set leans on a product name and jargon ('fidelity loop') and misses the most common variations a user would naturally say (e.g. 'match design to code', 'iterate on UI fidelity').	2 / 3
Distinctiveness Conflict Risk	The design-to-code niche is fairly specific, but 'design iteration' is a broad term that could overlap with other design-systems skills, so it is 'somewhat specific but could still overlap' rather than a clearly distinct, low-conflict trigger set.	2 / 3
	Total	8 / 12 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: nexu-io/open-design
Path: skills/stitch-loop/SKILL.md
Commit: a2ba3ee

Reviewed: 2 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.