progress-dashboard
Publishes a one-page stakeholder progress dashboard for any git project — where it stands, what's waiting on people, whether it's stalled — from whatever…
Skill by Busy-Office
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/busy-office-progress-dashboard
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, design, data, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/progress-dashboard
What's inside
- Skill name
progress-dashboard- Description
- Publishes a one-page stakeholder progress dashboard for any git project — where it stands, what's waiting on people, whether it's stalled — from whatever records the project keeps (ADRs, decision notes, design graphs, roadmaps, gates, backlogs, changelogs, PRs). Use whenever someone asks how a project is going, wants a status update or progress report, asks "where are we", "what's blocked", "what's waiting on me/us", "what's left", or wants an existing dashboard refreshed.
- License
- MIT
- Instructions
- 134 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (32)
- SKILL.md
- HUMAN-GATES-LOG.md · fixtures/frontmatter-dec/docs/
- ROADMAP.md · fixtures/frontmatter-dec/docs/
- expect.json · fixtures/frontmatter-dec/
- DEC-012 Platform Operator Flag.md · fixtures/frontmatter-dec/vault/wiki/decisions/
- DEC-013 Open Thing.md · fixtures/frontmatter-dec/vault/wiki/decisions/
- ROADMAP.md · fixtures/generated-roadmap/docs/
- expect.json · fixtures/generated-roadmap/
- BACKLOG.md · fixtures/line-graph/docs/
- DESIGN-GRAPH.md · fixtures/line-graph/docs/
- HUMAN-GATES-LOG.md · fixtures/line-graph/docs/
- LOOP-STATUS.md · fixtures/line-graph/docs/
- MANUAL-ACTIONS.md · fixtures/line-graph/docs/
- expect.json · fixtures/line-graph/
- CHANGELOG.md · fixtures/prose-adr/
- ROADMAP.md · fixtures/prose-adr/
- ADR-0001-platform.md · fixtures/prose-adr/docs/decisions/
- ADR-0002-stack.md · fixtures/prose-adr/docs/decisions/
- ADR-0003-parked.md · fixtures/prose-adr/docs/decisions/
- README.md · fixtures/prose-adr/docs/decisions/
- TEMPLATE.md · fixtures/prose-adr/docs/decisions/
- decisions.yaml · fixtures/prose-adr/docs/decisions/session-2026-09-07/
- expect.json · fixtures/prose-adr/
- ROADMAP.md · fixtures/prose-roadmap/
- STATUS.md · fixtures/prose-roadmap/
- expect.json · fixtures/prose-roadmap/
- DESIGN-GRAPH.md · fixtures/table-graph/
- PROJECT-PLAN.md · fixtures/table-graph/
- SESSIONS.md · fixtures/table-graph/
- expect.json · fixtures/table-graph/
- manifest.md · references/
- collect.mjs · scripts/
SKILL.md (first lines)
# Progress Dashboard A project's real status is scattered: decisions in one place, phases in another, sign-offs in a third, PRs on GitHub. This skill collapses that into one page for a **stakeholder** — someone who needs the answer, not the inventory. The first screen answers three things: where the project is, what is waiting on a human, and whether it has stalled. Everything else is one click away. ## Workflow ### 1. Collect ```bash node <skill-dir>/scripts/collect.mjs <repo-path> > <scratchpad>/progress-state.json ``` Flags: `--fast` (local files only: no GitHub, no per-file git history, no refresh command — ~0.2 s), `--no-gh` (skip GitHub only), `--profile` (timings to stderr), `--json-only` (no stderr notes), `--self-test` (run the fixtures instead of a repo). The collector is read-only and runs out-of-process; it never touches the target project's files, dependencies, build or git state. Full mode costs about 0.2–0.4 s for local records plus ~1.5 s for GitHub. Use `--fast` for a quick check (someone asks "where are we?" mid-session), full mode for the published page. The collector reads `.claude/progress.json` in the target repo if present (see `references/manifest.md`), then shape-detects everything the manifest doesn't pin. It never runs a command the manifest didn't declare. Output is one JSON blob; read it whole before writing anything. The floor is a git repo. If `warnings` says it isn't one, stop and say so. If it is a git repo with no decision or roadmap records, publish a short page from git and GitHub alone and say plainly what the project doesn't record yet. ### 2. Judge The collector is deterministic; the judgement is yours. From the JSON, form: - **Headline** — a fixed skeleton, worded by you: `[current phase and what's left] ; [staleness verdict] with [N] items waiting on people.` Current phase is `phases[].current`. Verdict is `staleness.verdict` (active / quiet / stalled / dormant) — use the word. - **Waiting on humans** — `humanItems`, already ranked blocks-first (phase exit → decision → review/other) then by age, grouped by `owner`. A PR that is green and unmerged is a human item; so is an open gate, an open question with an owner, an unchecked manual action. - **Phases** — `phases[]` with criteria counts where the roadmap has exit criteria; where it doesn't, show status only. Never invent a percentage. - **What's quietly wrong** — `notes.dataQuality`: vocabulary the collector couldn't map, a phase referenced but not defined, a decision file with no status. These go on the page. `notes.config` (declare X in the manifest) goes to the terminal, not the page. Where the data supports a conclusion, state it. "P2 is one merge away" beats a bar the reader has to interpret. ### 3. Publish Load `artifact-design`, then write the page and publish with the Artifact tool. Stable path: `<scratchpad>/<project>-progress.html`. If a dashboard for this project already exists, find it with `action: "list"`, read it, and publish with its `url` so the stakeholder's link keeps working. ### 4. Optionally mirror to Notion If the manifest has a `notion` key, or the person asks for the status "in Notion", update that page after publishing the artifact — don't create a second source of truth. Use the Notion connector (`notion-fetch` the page, then `notion-update-page`) and write only the first screen: the headline, both stamps, "Waiting on humans" grouped by owner, and the phase list, each as plain Notion blocks, followed by a link to the artifact for the full page. Replace the previous mirror's content rather than appending, so the Notion page always reads as the current state. If no `notion` key exists and the person asks, create the page as a private draft, put its URL in the manifest, and say where it lives. ## The page Order and density are the design. First screen, always open: 1. **Masthead** — project, branch/sha, phase N of M; the headline; a two-line lede; both stamps (*generated at* and *last moved*, date and time, with the age in words). 2. **Waiting on humans** — grouped by owner, severity stripe by what it blocks, age on the right. Omit the section if empty; never show it hollow. 3. **Phases** — one row per phase, current one marked; criteria bar when criteria exist, status pill otherwise. Collapsed under their heading with a count (native `<details>`, no script): 4. **Decisions** — tiles by canonical status; recent and unsettled rows; deliberate absences shown as decisions with replacement and reopen trigger. 5. **In flight** — PRs with CI state and what they relate to; issue counts. 6. **Momentum** — last 14 days from git log, changelog and session notes, newest first, each line tagged with its source; a one-line cadence note. 7. **Capabilities** — only when the manifest opts in: agents, skills, AGENTS.md, with "new since last release". 8. **Sources** — every source with how it was found (manifest / shape / origin / default), its shape, and when it last changed; then the data-quality notes. An empty section is omitted, not collapsed. Semantic colour is for state (good / warning / critical), separate from the accent. ## Judgement calls - **Closed is closed.** Struck-through nodes, `Status: CLOSED`, checked boxes, superseded ADRs — respect them. A resolved item shown as pending is worse than one omitted. - **Absence is a decision.** A `DA-` node or any record with a replacement and reopen trigger is deliberate. Present it under Decisions, never as a gap. - **Generated files are renderings.** If the collector reports a source as generated (prime-cm style), show its generation date and don't claim it is fresher than that.
From Busy-Office/ai-skills, MIT.
More from Busy-Office/ai-skills
- 0.570score
jev
Busy-Office
Get a fast, typed second opinion (yes/no, choice, or score with probabilities) from jev-ai.pro before acting.
SkillAFreshness: AClaude Code +9 - 0.566score
loop-doctor
Busy-Office
Reviews — and, when asked, sets up — a project's autonomous, scheduled loop — the routine that wakes on a cadence (cron, launchd, a /loop interval, a cloud…
SkillAFreshness: AClaude Code +9 - 0.566score
graph-engineer
Busy-Office
Turns a linear agent, prompt chain, or orchestration script into an engineered graph — nodes with contracts, edges that carry data, fan-out/fan-in only where…
SkillAFreshness: AClaude Code +9 - 0.529score
app-icon-generator
Busy-Office
Generate a complete app icon set — iOS AppIcon.appiconset, Android launcher + adaptive icons, and web favicons/PWA manifest icons — from a single square…
SkillBFreshness: BClaude Code +9 - 0.529score
requeue
Busy-Office
Reviews and reorders the queue an engineering loop or team works from — backlog, roadmap, todo, milestones — so the next work picked up is the right work.
SkillBFreshness: BClaude Code +9 - 0.525score
loop-economist
Busy-Office
Reviews what an engineering loop's actual runs cost and produced — reading Claude Code session transcripts and git history rather than the loop's documents.
SkillBFreshness: BClaude Code +9 - 0.525score
solo-flow
Busy-Office
Prescribes and audits a trunk-based git workflow for a repository with exactly one writer — a single autonomous agent or loop, with no other humans committing.
SkillBFreshness: BClaude Code +9 - 0.283score
wake-weight
Busy-Office
No description from the publisher yet.
SkillBFreshness: BClaude Code +9 - 0.283score
sharpen-intent
Busy-Office
No description from the publisher yet.
SkillBFreshness: BClaude Code +9 - 0.283score
loop-atlas
Busy-Office
No description from the publisher yet.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 20, 2026 (1 month ago)
- Last release
- Sep 8, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/Busy-Office/ai-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Busy-Office
- 0.570score
jev
Busy-Office
Get a fast, typed second opinion (yes/no, choice, or score with probabilities) from jev-ai.pro before acting.
SkillAFreshness: AClaude Code +9 - 0.566score
loop-doctor
Busy-Office
Reviews — and, when asked, sets up — a project's autonomous, scheduled loop — the routine that wakes on a cadence (cron, launchd, a /loop interval, a cloud…
SkillAFreshness: AClaude Code +9 - 0.566score
graph-engineer
Busy-Office
Turns a linear agent, prompt chain, or orchestration script into an engineered graph — nodes with contracts, edges that carry data, fan-out/fan-in only where…
SkillAFreshness: AClaude Code +9 - 0.529score
app-icon-generator
Busy-Office
Generate a complete app icon set — iOS AppIcon.appiconset, Android launcher + adaptive icons, and web favicons/PWA manifest icons — from a single square…
SkillBFreshness: BClaude Code +9 - 0.529score
requeue
Busy-Office
Reviews and reorders the queue an engineering loop or team works from — backlog, roadmap, todo, milestones — so the next work picked up is the right work.
SkillBFreshness: BClaude Code +9 - 0.525score
loop-economist
Busy-Office
Reviews what an engineering loop's actual runs cost and produced — reading Claude Code session transcripts and git history rather than the loop's documents.
SkillBFreshness: BClaude Code +9
Cite this page
progress-dashboard skill. Agent Mart. https://agentmart.dev/skills/busy-office-progress-dashboard (last updated 2026-10-24)