chart-interpretation
Skill0 starsChanged 9 days agoMITView on GitHub
Read a chart someone else made — image, screenshot, HTML, or dashboard — and extract insights, patterns, anomalies, and misleading framing. The reverse of authoring. Use when handed a chart to analyze, when checking whether a chart's visuals support its claim, or when a figure needs a written narrative.
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- fabioc-aloha/alex-cognitive-architectureedited, first seen Apr 26, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- fabioc-aloha/alex-cognitive-architecture★ 089% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository66 other files it gives its agents.
- AGENTS.mdAGENTS.md
- skills/bootstrap-project/resources/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- skills/bootstrap-project/resources/CLAUDE.mdCLAUDE.md
- act-tenetsSkill
- adversarial-reviewSkill
- annotate-screenshotSkill
- anti-hallucinationSkill
- ascii-chartSkill
- assess-brainSkill
- big-ideaSkill
- bootstrap-coreSkill
- bootstrap-projectSkill
- browser-toolsSkill
- chart-big-ideaSkill
- chart-vocabularySkill
- code-reviewSkill
- communication-craftSkill
- compile-brainSkill
- convertSkill
- corpus-qa-sweepSkill
- critical-thinkingSkill
- deep-reviewSkill
- doc-hygieneSkill
- docs-shellSkill
- docx-to-mdSkill
- ethical-reasoningSkill
- evaluate-before-adoptingSkill
- figure-generatorSkill
- flint-chartSkill
- flint-interactionsSkill
- flint-themeSkill
- git-workflowSkill
- html-to-mdSkill
- humanizerSkill
- install-visual-companionsSkill
- lint-clean-markdownSkill
- markdown-mermaidSkill
- markdown-sanitization-chainSkill
- md-to-emlSkill
- md-to-htmlSkill
- md-to-txtSkill
- md-to-wordSkill
- meditationSkill
- mutation-testingSkill
- planSkill
- platform-awarenessSkill
- print-svg-style-guideSkill
- proactive-awarenessSkill
- problem-framing-auditSkill
- project-capability-authoringSkill
- project-statusSkill
- render-chartSkill
- render-verifySkill
- rich-emailSkill
- risk-analysisSkill
- security-and-hardeningSkill
- setup-dependenciesSkill
- setup-enterprise-stackSkill
- spikeSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 72
Check every chart for these deceptive patterns:
- Distinctive: line 88
Always read these elements before interpreting the data:
- Distinctive: line 153
After interpreting, suggest what would strengthen the analysis:
- Very specific: line 165
Use the CSAR loop from chart-vocabulary Module 2 for structured chart reading:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Chart Interpretation
The reverse of chart authoring. Instead of data to chart, this reads chart to insight to narrative. It works on charts you did not make: screenshots, images, HTML, dashboard exports.
The cardinal rule: read what the chart says, then read what it doesn't say. The visible data points tell one story; the missing context, truncated axes, and suppressed categories tell another.
Composes with chart-vocabulary for chart-type
knowledge and critical-thinking for the
evidence-quality and bias checks that Module 4 applies to visuals.
Module 1: Chart Type Recognition
Identify the chart type to determine the correct reading strategy.
| Chart Type | Key Visual Features | Reading Strategy |
| Bar / Column | Rectangular bars, one axis categorical | Compare bar lengths, check sort order |
| Horizontal Bar | Bars extend left-to-right | Rank comparison, read labels first |
| Line | Connected points over axis | Follow trend direction, find inflections |
| Area | Filled region under line | Volume over time, stacking if multiple |
| Pie / Donut | Circular segments | Part-to-whole, count segments, check % |
| Scatter | Points in x-y space | Look for clusters, outliers, trend line |
| Bubble | Scatter with size encoding | Three dimensions: x, y, size |
| Histogram | Bars touching, x is continuous | Distribution shape, skew, outliers |
| Heatmap | Color grid | Pattern density, row-column relationships |
| Treemap | Nested rectangles | Hierarchical proportions |
| Sankey | Flow ribbons between stages | Volume flow, biggest paths |
| Box Plot | Box + whiskers | Median, IQR, outlier dots |
| Network | Nodes + edges | Clusters, hubs, isolates |
| Violin | Mirrored density curves | Distribution shape + density |
Module 2: Visual Decoding
Extract data from visual encodings:
| Encoding | What to Read | Precision Level |
| Position (axis) | Exact values from gridlines/labels | High (if labeled) |
| Length (bar) | Relative magnitude between items | High |
| Color hue | Category membership | Categorical only |
| Color intensity | Value magnitude in sequential scheme | Medium |
| Size (area) | Third variable (bubble, treemap) | Low (area perception is poor) |
| Angle (pie) | Proportion (poor human accuracy) | Low |
| Slope (line) | Rate of change | Medium |
Module 3: Pattern Detection
| Pattern | What to Look For | Significance |
| Trend | Consistent upward/downward direction | Growth, decline, momentum |
| Inflection | Direction change point | Market shift, intervention effect |
| Plateau | Flat region after growth/decline | Saturation, stabilization |
| Cluster | Groups of points in scatter/network | Natural segments, sub-populations |
| Outlier | Points far from the main group | Anomaly, error, or special case |
| Periodicity | Repeating pattern at intervals | Seasonality, weekly cycle |
| Gap | Missing data or discontinuity | Data quality issue or deliberate omission |
| Skew | Asymmetric distribution shape | Non-normal population, concentration |
| Dominance | One item >> all others | Power law, market leader, outlier |
Module 4: Misleading Visual Detection
Check every chart for these deceptive patterns:
| Deception | How to Detect | Actual Impact |
| Non-zero baseline | Y-axis starts above 0 | Exaggerates differences (sometimes 2-5x) |
| Truncated axis | Axis range excludes data or starts mid-range | Hides context, magnifies small changes |
| Dual axes | Two Y-axes with different scales | Implies correlation where none may exist |
| 3D effects | Perspective distortion on bars/pies | Area comparison becomes inaccurate |
| Cherry-picked range | Time window starts/ends at convenient point | Hides contrary trend outside window |
| Suppressed categories | "Other" aggregates significant items | Hides important segments |
| Area distortion | Variable-width bars, non-proportional icons | Size doesn't match value |
| Missing denominator | Percentages without base size | 50% of 10 ≠ 50% of 10,000 |
| Reversed axis | Values increase downward or rightward | Readers misread direction |
Module 5: Structural Element Reading
Always read these elements before interpreting the data:
| Element | What to Extract | If Missing |
| Title | Author's intended takeaway | Chart lacks stated purpose |
| Subtitle | Time range, filter condition, context | Context must be inferred |
| Axes labels | What variables are plotted | Interpretation becomes guesswork |
| Legend | Category-to-color mapping | Color meaning unclear |
| Annotations | Author-highlighted insights | No guided reading |
| Data source | Where the data came from | Credibility unknown |
| Date/time | When data was collected/reported | Freshness unknown |
Module 6: Narrative Extraction
Convert visual observations into prose at three audience levels:
Executive Summary (30 seconds)
3 bullets maximum:
• [Primary insight — the main takeaway]
• [Supporting evidence — the strongest proof point]
• [Recommendation or implication — what to do about it]Detailed Analysis (2-3 minutes)
The chart shows [chart type] plotting [X variable] against [Y variable]
for [time range / scope].
Primary finding: [Main pattern or insight with specific numbers]
Supporting observations:
- [Pattern 1 with evidence]
- [Pattern 2 with evidence]
- [Anomaly or exception worth noting]
Context and caveats:
- [What the chart doesn't show]
- [Potential biases or limitations]
- [Comparison to benchmarks if available]Talking Points (presenter-ready)
"What you're seeing here is [explain the main pattern in plain language]."
"The key number to focus on is [highlight], which tells us [implication]."
"What's interesting is [surprise or anomaly] — this suggests [hypothesis]."
"The action item here is [recommendation]."Module 7: Confidence Rating
Rate interpretation confidence honestly:
| Level | When | Signal to User |
| High | Clear labels, clean data, familiar chart type | "The chart clearly shows..." |
| Medium | Some inference needed (unlabeled, partial data) | "Based on visual estimation..." |
| Low | Ambiguous visual, missing context, blurry image | "This appears to show, but verify..." |
Module 8: Follow-Up Recommendations
After interpreting, suggest what would strengthen the analysis:
| Suggestion Type | Example |
| Missing variable | "Add cost data to see if revenue growth is profitable" |
| Time extension | "Extend to 24 months to confirm the seasonal pattern" |
| Segmentation | "Break this down by region to check for Simpson's Paradox" |
| Alternative chart | "A scatter plot would better show the correlation" |
| Baseline addition | "Add a target line to show performance vs. plan" |
Module 9: CSAR Loop Integration
Use the CSAR loop from chart-vocabulary Module 2
for structured chart reading:
| Phase | Action |
| Clarify | What chart type? What variables? What time range? |
| Summarize | State the main finding in one sentence |
| Act | Extract specific data points, patterns, anomalies |
| Reflect | What's missing? What would I want to see next? |
Anti-Patterns
| Anti-Pattern | Problem | Fix |
| Describing, not interpreting | "This is a bar chart" (no insight) | Say what the bars MEAN, not what they ARE |
| Ignoring the title | Missing the author's intended message | Read title first -- it's the thesis |
| Over-precision from visual | "Revenue is exactly $4,237,892" | Estimate from visual: "roughly $4.2M" |
| Missing bias check | Accepting the chart at face value | Always scan for misleading elements |
| Single-lens reading | Only one interpretation offered | Provide primary + alternative reading |
Related Skills
chart-vocabulary— chart types by communication goal, and the CSAR loop Module 9 usescritical-thinking— Module 4's deception checks are evidence-quality and bias detection applied to a pictureflint-chart— the authoring direction. Read a chart here, make one thererender-verify— checks a chart you rendered says what you meant. This skill reads charts you did not make
Would Revise If
Revise if interpretations consistently state what a chart is rather than what it means, or if the confidence ratings in Module 7 prove uncalibrated against later verification of the underlying data.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

