ap-dashboard

name	ap-dashboard
description	Generate sprint run dashboard reports. Use when the user says 'generate dashboard', 'show sprint status', or 'autopilot report'.

ap-dashboard

Overview

This skill generates sprint run dashboard reports for the BMad Autopilot module. It reads sprint-status.yaml, cross-references actual git branch state, and produces a self-contained report showing what ran, what passed, what failed, and where things stand.

Act as a clear, accurate reporter. The dashboard is the user's window into what the autopilot did -- accuracy over aesthetics.

Capabilities:

• Generate Report -- Full sprint dashboard as self-contained HTML (default) or markdown. Includes per-story status cards, progress tracking, branch state, and file change counts.
• Summary -- Quick one-line-per-story text output for Telegram messages or terminal checks.

Args: [sprint-status-path] (optional, defaults to {implementation_artifacts}/sprint-status.yaml), --format html|markdown, --summary (summary mode).

On Activation

Load available config from {project-root}/_bmad/config.yaml and {project-root}/_bmad/config.user.yaml (root level and ap section). If config is missing, let the user know ap-setup can configure the module at any time. Use sensible defaults for anything not configured.

Resolve these config values (defaults in parens):

• dashboard_format (html) -- output format
• project_label (project name from sprint-status.yaml) -- display name in reports
• implementation_artifacts ({project-root}/_bmad-output/implementation-artifacts) -- where sprint-status.yaml lives

If --summary is passed or the user asks for a quick status, produce Summary output only.

Generate Report

1. Run the data gathering script to get structured sprint data:

   uv run scripts/gather-sprint-data.py <sprint-status-path> --verbose
   

   If the script cannot execute, perform equivalent data gathering by reading sprint-status.yaml directly and checking git branches via git branch --list.

2. Using the JSON output, generate the dashboard in the configured format.

HTML report requirements:

• Self-contained single file with inline CSS (no external dependencies)
• Clean, modern design with a muted color palette
• Project name and generation timestamp in the header
• Overall progress bar (done/total stories)
• Per-epic sections with story cards showing:
  • Story ID and status (color-coded: done=green, in-progress=blue, review=amber, backlog=gray, ready-for-dev=teal)
  • Branch name (linked if possible) or "No branch" for backlog stories
  • Files changed count (if branch exists)
  • Whether a story file exists
• Summary statistics at the bottom

Markdown report requirements:

• Clean table-based layout, readable in terminal via cat
• Same information as HTML: progress, per-epic tables, summary
• Status indicators using text labels (no emoji unless user configured them)

3. Write the report to {implementation_artifacts}/autopilot-dashboard.html (or .md). Inform the user of the output path.

Summary

Produce a compact text block with one line per story that has activity (skip pure backlog). Format:

[Project Label] Sprint Status
Epic 1: 2/5 done
  1-1 first-story .......... done
  1-2 second-story ......... in-progress (branch: story/1-2, 14 files)
  1-3 third-story .......... review

This format is designed for Telegram messages and terminal output. Return it as text, do not write to a file unless asked.