agentleFS
datadog-labsdatadog-labs/claude-code-plugin/skills/visualizations/SKILL.md

visualizations

Skill18 starsChanged 6 months agoApache-2.0View on GitHub

Use ONLY when the user explicitly asks to see, show, graph, visualize, chart, or plot data, or when another loaded Datadog data-domain skill directs you to visualization guidance. This is the plugin's local copy of the `datadog/visualizations` server skill; prefer it over fetching the server version.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. nateherkai/a-bunch-of-skills★ 41same name

    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 repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 2 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this guide when the user asks to see, show, graph, visualize, or chart data.

    line 8
  2. Very specific:

    Check this first. The tools below come from the Datadog MCP server's visualizations toolset, which is not enabled by default on every …

    line 12
  3. Very specific:

    If other Datadog MCP tools are available, the visualizations toolset is disabled.

    line 14
  4. Very specific:

    If no Datadog MCP tools are available, the server is not set up or not working.

    line 15
  5. Very specific:

    Key rule: If the user says "show me", "graph", "visualize", "chart", or "plot" — always use get_widget or visualize_tabular_data.

    line 26
  6. Very specific:

    Metrics visualization: Call get_widget with a widget_definition containing the metric query:

    line 42
  7. Very specific:

    If a schema-correct retry still fails, call ask_widget_expert for targeted debugging.

    line 74
  8. Distinctive:

    Do NOT only return a markdown table when the user asked for a visualization

    line 86

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Visualization Guide

Use this guide when the user asks to see, show, graph, visualize, or chart data.

When the Visualization Tools Are Unavailable

Check this first. The tools below come from the Datadog MCP server's visualizations toolset, which is not enabled by default on every client. If neither get_widget nor visualize_tabular_data is available, do not call them and do not guess other tool names:

  • If other Datadog MCP tools are available, the visualizations toolset is disabled. Tell the user that Datadog charts are not enabled and suggest running /ddtoolsets to enable visualizations.
  • If no Datadog MCP tools are available, the server is not set up or not working. Suggest running /ddsetup.

Then answer with the data you can retrieve, as text or a table. In this case only, the visualization rules below do not apply.

Tool Selection

User intentToolWhen to use
Show a chart of Datadog dataget_widgetTimeseries, toplists, query tables, query values, heatmaps, distributions — any standard Datadog widget
Render tabular data as a chartvisualize_tabular_dataWhen you have rows/columns that aren't directly queryable via a Datadog widget definition

Key rule: If the user says "show me", "graph", "visualize", "chart", or "plot" — always use get_widget or visualize_tabular_data. Do NOT only return a markdown table or text summary for these requests.

Chart Type Selection

Data shapeRecommended chart type
Metric over timetimeseries via get_widget
Grouped data / proportional breakdown / pie chartsunburst via visualize_tabular_data
Ranked list / bar chart (top N by value)toplist via get_widget or visualize_tabular_data
Nested / hierarchical breakdowntreemap via visualize_tabular_data
Single number / KPIquery_value via get_widget (supports background timeseries and delta indicator)
Change over time across groupschange via get_widget
Distribution of valuesdistribution via get_widget

Common Patterns

Metrics visualization: Call get_widget with a widget_definition containing the metric query:

{
  "type": "timeseries",
  "requests": [
    {
      "response_format": "timeseries",
      "queries": [{ "name": "query1", "data_source": "metrics", "query": "avg:system.cpu.user{*}" }]
    }
  ]
}

Log/span analytics as chart: Build the query directly in a get_widget call (data_source: "logs", "spans", etc.) — this keeps the widget exportable to Datadog. Use visualize_tabular_data only as a last resort when the query isn't representable in a widget definition.

Calling get_widget

The three input modes are mutually exclusive:

  • Direct definition (ad hoc): write a new widget from scratch by providing a widget_definition object. See Building widget_definition for the schema-lookup workflow.
  • By ID (dashboard_id + widget_id): fetch data for an existing dashboard widget you found via search_datadog_dashboards.
  • By URL: provide a Datadog widget_url (a link copied from the web UI). Supports share links (/s/...), short links (/u/...), and dashboard links (/dashboard/... with a focused or fullscreen widget).

Pass the widget object as widget_definition, not definition (wrong key is the most common 400).

Building widget_definition

For ad-hoc widgets (widget_definition mode), each widget type has a unique schema. Learn the schema by calling get_widget_reference with the list of widget types you are considering building, then construct widget_definition from the returned schema and pass it to get_widget.

For existing dashboard widgets, prefer search_datadog_dashboards → get_widget since the saved definition is already valid.

If a schema-correct retry still fails, call ask_widget_expert for targeted debugging.

get_widget response

Responses include METADATA (widget_type, query metadata) and DATA sections. The DATA shape varies by widget_type; see data_shape and the render_summary.text field in each response for what came back.

When get_widget fails

Error responses include a [RECOVERY] or [EMPTY WIDGET] action list — follow it. Focus retries on schema-level corrections; cosmetic changes (title, display_type) rarely resolve validation errors. If you have not already called get_widget_reference for the affected widget_type this turn, call it, then retry. After 2 failed attempts on the same visualization, explain the blocker to the user in text instead of looping.

What NOT to Do

  • Do NOT only return a markdown table when the user asked for a visualization

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.