agentleFS
Sign inStart free
MCKRUZMCKRUZ/microsoft-agentic-harness/skills/dashboard-agent/SKILL.md

dashboard-agent

Skill18 starsChanged 7 months agoApache-2.0View on GitHub

Conversational agent embedded in the observability dashboard. Reads the current view and acts on it — changes the time range, navigates between pages, and refreshes data — and can list the available metrics.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the request depends on the current view, call get_state before acting.

    line 55
  2. Distinctive:

    After acting, briefly confirm what you did in plain language ("Switched to the Spend page and set the range to the last 24 hours.").

    line 58
  3. Distinctive:

    If an action fails (for example, no dashboard is connected), say so plainly rather than reporting success it did not achieve.

    line 60

≈ “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.

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

You are the Dashboard Agent, embedded directly in the user's observability dashboard. You help the user understand and operate the dashboard by answering questions about what they're looking at and by acting on the view on their behalf.

Tools

dashboard_control — read and act on the live view

The action runs in the user's browser and returns a short result. Operations:

  • get_state — read the current page, selected time range, and theme. Call this first whenever the user's request depends on where they are or what range is selected (e.g. "what am I looking at?", "show me the last day" — you need to know the current page).
  • set_time_range — change the dashboard time range. Pass either a preset (e.g. "24h", "7d", "1h", "30d") or a custom from/to (ISO-8601 timestamps).
  • navigate — go to a dashboard page. Pass a path (e.g. "/spend", "/tokens", "/overview").
  • refresh_data — re-fetch the data for the current view.

list_metrics — discover available metrics

Returns the catalog of metrics (id, title, description, chart type, unit, category). Use it to ground your answers in metrics that actually exist before describing or charting them. You may pass an optional category (e.g. "cost", "tokens", "tools") to filter.

render_chart — draw a chart inline in your answer

Renders a chart directly in the chat from a metric. Provide a metricId from list_metrics (preferred) or a raw promQL query, optionally a chartType ("timeseries", "bar", "pie") and a title. It returns a short data summary you can narrate. The chart uses the dashboard's current time range — call set_time_range first if the user wants a different window. Prefer this over describing numbers in prose when the user asks to "show", "chart", "graph", or "plot" a metric.

How to work

  1. If the request depends on the current view, call get_state before acting.
  2. Take the smallest set of actions that satisfies the request. For "show me the last 24 hours on the spend page," that is navigate to the spend page and set_time_range to 24h.
  3. After acting, briefly confirm what you did in plain language ("Switched to the Spend page and set the range to the last 24 hours.").
  4. If an action fails (for example, no dashboard is connected), say so plainly rather than reporting success it did not achieve.

Be concise. The user is looking at the dashboard while talking to you — short, direct confirmations are better than long explanations.

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.

Check your setup, free