rill-canvas

Detailed instructions and examples for developing canvas dashboard resources in Rill

Quality

60%

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/rill-canvas/SKILL.md

Quality

Content

71%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 and accurate, with complete YAML examples for every component and precise syntax rules, and it includes a clear recommended build sequence. Its weaknesses are redundancy across near-duplicate chart examples and a monolithic structure that inlines per-component reference material that belongs in separate reference files.

Suggestions

Split per-component reference details (e.g. component-type specs, custom-chart/Vega-Lite rules, field configuration) into references/ files and keep SKILL.md as an overview with well-signaled one-level-deep links.

Consolidate near-duplicate examples (line_chart's three variants, stacked_bar vs stacked_bar_normalized, and the Complete Example) to trim token cost without losing coverage.

Add a brief validation step to the composition workflow, such as verifying that all field names exist on the referenced metrics view and that the YAML parses, before finishing a dashboard.

Dimension	Reasoning	Score
Conciseness	The body is dense with Rill-specific syntax and avoids explaining concepts Claude already knows, but contains substantial redundancy: three near-identical line_chart variants, stacked_bar/stacked_bar_normalized near-duplicates, and a Complete Example that re-demonstrates components already shown.	3 / 5
Actionability	Every component type ships a full copy-paste-ready YAML block, with exact syntax rules for defaults.filters and metrics_sql, complete Vega-Lite JSON specs, and a fully executable end-to-end dashboard example.	5 / 5
Workflow Clarity	'Dashboard Composition Best Practices' provides an explicit numbered build sequence plus chart-type selection guidance and guardrails ('Never use the time dimension in leaderboard dimensions'), but there are no explicit validation/verification steps such as checking field names against the metrics view.	4 / 5
Progressive Disclosure	No bundle files exist; all per-component reference details, custom-chart/Vega-Lite rules, and field configuration live inline in one ~1450-line SKILL.md. Headers and navigation are clear, but content that clearly belongs in separate reference files is fully inlined.	3 / 5
	Total	15 / 20 Passed

Description

48%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 correctly identifies a specific niche (Rill canvas dashboards) but reads like a table of contents rather than a capability statement. It lacks any trigger guidance ('Use when...'), lists no concrete actions, and omits natural synonyms like 'report' or 'executive summary'.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when building fixed reports, executive summaries, or multi-visual dashboards in Rill, or when the user mentions canvas dashboards or Rill reports.'

Enumerate 2-4 concrete capabilities instead of 'Detailed instructions and examples', e.g. 'Build canvas dashboards with KPI grids, line/bar/heatmap charts, leaderboards, and filters, plus custom Vega-Lite charts.'

Add distinguishing trigger terms against the sibling explore-dashboard skill (e.g. 'fixed reports and executive summaries' vs 'explorative analysis').

Dimension	Reasoning	Score
Specificity	The description names the domain ('canvas dashboard resources in Rill') but the action is generic ('Detailed instructions and examples for developing') with no enumerated concrete capabilities such as building KPI grids, charts, or filters.	2 / 5
Completeness	There is a reasonably clear 'what' but no 'when'/'Use when' clause of any kind, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Contains genuine natural keywords ('Rill', 'canvas dashboard', 'dashboard resources') but misses common variations users would say, such as 'report', 'executive summary', 'KPI dashboard', or 'Rill dashboard'.	3 / 5
Distinctiveness Conflict Risk	'Canvas dashboard resources in Rill' is a distinct niche with minimal conflict risk against unrelated skills, though it does not disambiguate against a closely related sibling such as a Rill explore-dashboard skill.	4 / 5
	Total	12 / 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
skill_md_line_count	SKILL.md is long (1448 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: rilldata/agent-skills
Path: skills/rill-canvas/SKILL.md
Commit: 1307e7f

Reviewed: 7 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.