- Home
- Skills
- Research & Analysis
- chart _quality _reviewer
chart _quality _reviewer
Audit chart metadata for accessibility, scale accuracy, and publication standards in research and business reports.
Free
Works with the AI Agents you already use
chart _quality _reviewer
Example session with this skill installed
Review my chart: bar chart, X is Month, Y is Revenue in USD (0 to 10k), legend has Actual and Forecast, palette is red and green, for a business_report. Caption: Q3 performance.
- Read your context and instructions
- Compiled the chart _quality _reviewer
I have reviewed the metadata for your Q3 performance bar chart. The chart compares Actual versus Forecast revenue across months with a 0 to 10k USD scale.
Chart Quality Review Report
| Finding | Severity | Impact | Required Fix |
|---|---|---|---|
| Red-Green Palette | Critical Risk | Inaccessible to color-blind users (deuteranopia). | Change palette to a color-blind safe set like Viridis or blue-orange. |
| Missing X-axis Units | Minor Improvement | Ambiguity in time series representation. | Explicitly state if months are fiscal or calendar year. |
| Scale Validation | Passed | Linear scale starts at zero. | No action needed. |
Top 3 Priority Modifications
- Replace the Red-Green palette immediately to ensure the chart is accessible and professional.
- Standardize legend labels to ensure "Actual" and "Forecast" colors are distinct in greyscale.
- Verify tick intervals on the Y-axis to ensure 0-10k range has logical increments (e.g., 2k).
Disclaimer: This report is advisory. The final decision on figure presentation rests with the human author or editor.
Next steps
- Update the color palette to a high-contrast, accessible alternative.
- Add a specific year or fiscal period to the X-axis label.
- Re-run the review once the new color hex codes are selected.
Connects securely to your tools. The creator never sees your data.
About this skill
Poorly formatted charts can get research rejected or lead to bad business decisions. This skill audits chart metadata to ensure your visualizations meet rigorous publication and professional standards.
What it does
- Standard compliance checks for axis labels, units, and legend mapping.
- Scale validation identifies truncated or distorted scales that mislead viewers.
- Accessibility audit flags color-blind safety issues and palette choices.
- Scenario-specific grading adjusts rules for academic journals versus business reports.
- Actionable feedback delivers a graded list of fixes from critical risks to minor tweaks.
How it works
- Data ingestion: Provide a structured description of your chart including type, axes, units, and legend details.
- Schema normalization: The agent converts your description into a standard JSON format for analysis.
- Automated auditing: A deterministic script checks your data against a catalog of visualization rules.
- Expert synthesis: The agent reviews the script output and adds context-specific judgment.
- Report generation: You receive a structured review report with prioritized modifications.
Frameworks & tools
The skill utilizes a Python-based rule engine check_chart.py and structured JSON schemas to ensure objective auditing of visualization metadata.
Why this beats prompting it yourself
Generic prompts often miss subtle technical requirements like unit consistency or specific color-blind safety patterns. This tool uses a deterministic checker to catch every violation of formal charting rules without the hallucinations or inconsistencies of raw LLM analysis.
Use cases
- Audit figures before submitting to peer-reviewed academic journals.
- Verify business report charts for clarity and non-misleading scales.
- Ensure data visualizations are accessible to color-blind stakeholders.
- Create a pre-submission checklist for data science teams.
Known limitations
Does not process images or screenshots. Requires text-based metadata for the chart. Does not evaluate statistical validity or research methodology.
Cannot read image files or screenshots. Requires text or JSON descriptions of chart metadata. Does not write plotting code or perform statistical analysis.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed today
- Free with an Agensi account