v10/.github/scripts/project-report/prompt.md

You are generating the weekly progress report for Video.js 10.

Step 1: Determine date range

Calculate the most recent completed full week (Monday–Sunday). If today is Monday, that means last Mon–Sun. If today is any other day, it is the Monday–Sunday of the current or just-ended week.

The display label should show the full week: "Feb 23 – Mar 1, 2026". For GitHub search queries, use exactly that Monday–Sunday range.

Step 2: Collect data from GitHub

Use gh CLI commands. Batch GraphQL queries with aliases to stay within rate limits (5,000 points/hour).

Project board items

Fetch ALL items with manual cursor pagination. Do not use --paginate — it duplicates GraphQL results. Use this pattern:

CURSOR=""
ALL_ITEMS="[]"
while true; do
  if [ -z "$CURSOR" ]; then AFTER_ARG=""; else AFTER_ARG=", after: \"$CURSOR\""; fi
  RESULT=$(gh api graphql -f query='{ node(id: "PVT_kwDOADIolc4BHP_1") { ... on ProjectV2 { items(first: 100'"$AFTER_ARG"') { pageInfo { hasNextPage endCursor } nodes { status: fieldValueByName(name: "Status") { ... on ProjectV2ItemFieldSingleSelectValue { name } } points: fieldValueByName(name: "Story Points") { ... on ProjectV2ItemFieldNumberValue { number } } content { ... on Issue { number title state milestone { title } labels(first: 10) { nodes { name } } assignees(first: 5) { nodes { login } } } } } } } } }')
  ITEMS=$(echo "$RESULT" | jq '.data.node.items.nodes')
  ALL_ITEMS=$(echo "$ALL_ITEMS $ITEMS" | jq -s '.[0] + .[1]')
  HAS_NEXT=$(echo "$RESULT" | jq -r '.data.node.items.pageInfo.hasNextPage')
  CURSOR=$(echo "$RESULT" | jq -r '.data.node.items.pageInfo.endCursor')
  if [ "$HAS_NEXT" != "true" ]; then break; fi
done
echo "$ALL_ITEMS" > /tmp/project_items.json

Merged PRs and closed issues

gh pr list --repo videojs/v10 --state merged --search "merged:START..END" --limit 100 --json number,title,author
gh issue list --repo videojs/v10 --state closed --search "closed:START..END" --limit 100 --json number,title,closedAt

UC epic sub-issues

Fetch sub-issues via REST, then batch story points with aliased GraphQL:

gh api repos/videojs/v10/issues/{NUMBER}/sub_issues --jq '[.[] | {number, title, state}]'

Known UC epics:

Epic	Issue	Owner
UC-1 Core Playback UI	#489	Rahim
UC-2 Adaptive Streaming	#353	Wes
UC-3 Skins	#490	Sam
UC-4 Captions & Subtitles	#491	Rahim
UC-5 Accessibility	#492	Rahim
UC-8 Keyboard & Power-User	#494	Rahim

Check for new epics each week:

gh issue list --repo videojs/v10 --label epic --milestone Beta --state all --json number,title

Step 3: Calculate metrics

Apply these rules to the collected data:

• Filter to Beta milestone only
• Exclude items labeled epic — these are rollups that double-count SP
• Status values are case-sensitive: Done, In progress, Up next, Ready for review, Blocked
• SPF dominates (~66% of SP) — always compute a "Without SPF" breakdown

Workstream label mapping

Label	Workstream
spf	SPF
components	UI Components
skin	Skins
media	Media
a11y	Accessibility
Any label starting with docs	Docs & Guides
site	Site
compiler	Compiler
store, pkg:core, pkg:dom	Core / Store
Everything else	Other

Compute:

1. Velocity — SP and items completed THIS WEEK (closed in the date range)
2. Beta progress — all-time by status (Done, Active, Blocked, Unplanned)
3. Without SPF — same breakdown excluding spf-labeled items
4. By workstream — total, done, active, blocked per workstream
5. Team breakdown — this week's completed SP/items/PRs per assignee
6. UC epics — done SP / total SP per epic
7. Blocked items — anything with status "Blocked"
8. What's next — open items with status In progress, Up next, Ready for review

Step 4: Post to Slack

Write the report as a JSON file at /tmp/report.json matching this exact structure, then run the Slack posting script.

{
  "header": "Feb 23 – Mar 1, 2026",
  "velocity": {
    "sp_completed": 42,
    "items_completed": 8,
    "prs_merged": 12,
    "issues_closed": 8
  },
  "beta_progress": {
    "total_items": 150,
    "total_sp": 234,
    "by_status": [
      { "status": "Done", "sp": 89, "count": 45 },
      { "status": "In progress", "sp": 30, "count": 12 }
    ]
  },
  "without_spf": {
    "total_sp": 156,
    "done_sp": 67,
    "active_sp": 34,
    "blocked_sp": 5,
    "unplanned_sp": 50
  },
  "by_workstream": [
    { "name": "SPF", "total": 78, "done": 35, "active": 20, "blocked": 3 }
  ],
  "team_breakdown": [
    { "assignee": "rahim", "sp": 18, "items": 5, "prs": 4 }
  ],
  "uc_epics": [
    { "id": "UC-1", "name": "Core Playback UI", "owner": "Rahim", "done_sp": 10, "total_sp": 25 }
  ],
  "blocked_items": [
    { "number": 142, "title": "Waiting on external API docs" }
  ],
  "whats_next": [
    { "number": 200, "title": "Slider a11y audit", "status": "In progress", "assignees": ["rahim"], "sp": 5 }
  ],
  "ai_summary": "2-3 paragraph narrative. See writing guidelines below."
}

Then post it:

cat /tmp/report.json | node .github/scripts/project-report/post-to-slack.js

Writing style for ai_summary

• Direct, confident, friendly but not chatty
• Active voice, short sentences
• No filler: "In order to", "basically", "simply", "just", "very", "actually"
• No hedging: "might", "could", "perhaps"
• Use "we" and "our" for the project team
• Reference issues as #NUMBER, people as @name
• Cover: what shipped, velocity trends, risks/blockers, what to focus on next

Gotchas

• Status values are case-sensitive in jq. Verify with [.[].status.name] | unique.
• --paginate duplicates GraphQL results. Always use manual cursor pagination.
• Write jq to a file. Shell quoting of // (jq's alternative operator) breaks in bash inline strings. Save to /tmp/calc.jq and use jq -f.
• fieldValueByName needs aliases. Can't appear twice in one GraphQL selection. Use status: fieldValueByName(name: "Status") and points: fieldValueByName(name: "Story Points").