visualization-ops-skill

**Required for visualized output.** When the prompt involves news/briefings/market updates/rankings/reports/summaries/analyses/comparisons/selection/recommendations/roundups/retrospectives, or is expected to return 3+ structurable items, **this skill must be loaded before (or in the same turn as) the data-fetching tool**. Once loaded, **do not finish with plain markdown text**—you must render via JsxCreate or ChartRender. Typical examples: "search the latest news about Iran", "compare Obsidian/Notion/Logseq", "recommend a few books", "how is the stock market recently". **Not for**: single-fact Q&A, pure code output, adding/removing nodes inside the canvas (handled by the canvas sub-agent).

Quality

73%

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 ./apps/server/src/ai/builtin-skills/visualization-ops/en/SKILL.md

Quality

Content

57%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.

Highly actionable content with strong executable examples and a clear routing decision tree, undermined by contradictory JsxCreate failure-recovery instructions and bulk reference material inlined in a single long file with no external references.

Suggestions

Reconcile the two JsxCreate failure-recovery sections into one explicit validate→fix→retry loop so Claude gets a single unambiguous procedure.

Move the component whitelist and color-token catalog into a separate references file (e.g. references/components.md) and link to it from the body, slimming SKILL.md to an overview.

Trim the letter analogy and the named-model aside ('from models such as MiniMax') to reduce tokens without losing the core 'do not pre-announce render calls' guidance.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with concrete tables, decision tree, and examples, but includes unnecessary padding: the letter-writing analogy, a six-line play-by-play anti-pattern, the 'MiniMax' aside, and restating 'supports all ECharts chart types' after the inventory table already says it.	3 / 5
Actionability	Concrete, executable guidance throughout — complete ECharts option JSON for bar/pie, a wrong-vs-correct JSX example, parameter lists, a component whitelist, and semantic color tokens. Minor gaps: no end-to-end realistic card example combining tokens and whitelisted components.	4 / 5
Workflow Clarity	A clear decision tree and validation-correction loops are present, but the two failure-recovery sections directly contradict each other: one says 'Prefer re-calling JsxCreate with a full corrected content... Only use Edit... when the issue is clearly isolated', while the other says 'After a failure, fix via Edit—do not call JsxCreate again'.	3 / 5
Progressive Disclosure	Well-sectioned single file, but at ~180 lines with no bundle files all reference material (the 30+ component whitelist, the full color-token catalog) is inlined rather than split into separate reference files, and no external references are signaled.	3 / 5
	Total	13 / 20 Passed

Description

90%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, trigger-rich description that clearly states what the skill does and when to load it, with good boundary guidance. The main weakness is second-person voice ('you must render', 'do not finish'), which the rubric penalizes.

Suggestions

Rewrite in third person to avoid the voice penalty — e.g. 'Renders visualized output via JsxCreate or ChartRender' instead of 'you must render via...'.

Tighten the long enumeration of content types into a shorter representative list plus an 'etc.' to reduce length without losing trigger coverage.

Dimension	Reasoning	Score
Specificity	Names the domain (visualized output) and several concrete actions ('render via JsxCreate or ChartRender', 'loaded before... the data-fetching tool', 'do not finish with plain markdown text'), but second-person phrasing ('you must render') triggers the rubric's voice penalty, reducing this from 4 to 3.	3 / 5
Completeness	Explicitly answers both 'what' (render visualized output via JsxCreate/ChartRender instead of plain markdown) and 'when' (prompts involving the listed content types or 3+ structurable items), with concrete trigger phrases and a 'Not for' boundary.	5 / 5
Trigger Term Quality	Comprehensive natural-term coverage — 'news/briefings/market updates/rankings/reports/summaries/analyses/comparisons/selection/recommendations/roundups/retrospectives' plus real user phrasings like 'compare Obsidian/Notion/Logseq' and 'how is the stock market recently'.	5 / 5
Distinctiveness Conflict Risk	Clear niche tied to two named rendering tools, with an explicit 'Not for' exclusion (single-fact Q&A, pure code output, canvas nodes) that minimizes overlap with sibling skills.	5 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: itsablabla/openloaf-web
Path: apps/server/src/ai/builtin-skills/visualization-ops/en/SKILL.md
Commit: a1ab5be

Reviewed: 4 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.