Skip to content

MVPavan/via / .claude/skills/html-artifact

html-artifact

Verified

Use for requested standalone HTML documents or interactive explanations when that format materially helps. Skip agent-facing files, Markdown-native deliverables, and ordinary short answers.

references/preset-report.md

View on GitHub
# Preset: report

For plans, status updates, post-mortems, architecture docs, design specs, PR writeups, option comparisons, retrospectives — anywhere a reader expects to skim by section. If charts are the argument and prose only labels them, use `dashboard` instead.

## Structure

A report has three layers:

1. **Orientation at the top.** Title, one-line subtitle, and a row of KPI tiles. The reader should answer "what is this and is it current?" in five seconds.
2. **Numbered sections in the middle.** Four to seven sections, each with a clear scannable header. The reader should be able to jump to the section they care about.
3. **Source attribution at the bottom.** A small grey footer listing where the doc came from and when it was generated.

## Section order by report type

The order matters more than people realise — reports of the same type should feel similar. Pick from these defaults and only deviate with reason:

- **Implementation plan**: Milestones · Data flow · Mockups · Key code · Risks · Explicitly not doing · Open questions
- **Status report**: KPIs · Highlights · Shipped · In flight · Blocked/slipped · Asks · Sources
- **Post-mortem**: Summary · Timeline · Root cause · Customer impact · What worked · Follow-ups · Sources
- **Architecture doc**: Context · System diagram · Components · Data flow · Failure modes · Capacity
- **Design spec**: Problem · Constraints · Proposal · Alternatives · Open questions
- **PR writeup**: Motivation · Before/after · File tour by theme · Where to focus · Risks and testing · Open questions
- **Option comparison**: Framing · One column per option · Hard-metrics row · Recommendation

## What each section looks like

**Milestones and timelines** are rendered as visual timelines — a strip or vertical spine with real spacing — never a bulleted list; the reader needs to *see* the pace.

**"Asks"** in a status report is a visually separated block — calls to action get lost when intermixed. **"What worked"** sits beside what didn't in a post-mortem, and follow-ups carry owners *and* dates — commitments, not a wishlist.

**Before/after** is a real side-by-side, not "before: X, after: Y" prose. In a PR writeup the file tour is grouped by theme (plumbing / core logic / tests / docs), not alphabetically.

**Option comparison** columns share an identical internal structure — a metric missing from one column reads as a hidden weakness — and the artifact ends with an actual recommendation, not "it depends".

**Risks / open questions** are tables, not bullet lists. Three columns: the thing, its severity, and the mitigation. Severity uses the severity tag pill. Tables convey shape — "five mediums and one high" reads instantly from a column.

**Key code** is in fenced figures: a muted caption above naming the file path, the code below in monospace, and a small "Copy" button. Avoid pasting in long blocks — pick the 5-20 lines a reader would actually want to copy.

**Mockups** are sketched in HTML using the page's design language — not embedded screenshots. They should feel like part of the doc, not pasted in. Use the surface, border, and muted-text colors for a "sketch" feel.

**Data flow diagrams** are inline SVG (rules in `svg-craft.md`): boxes for stages, arrows between them, dashed lines for asynchronous paths. Required once the system has more than two components — prose cannot carry topology.

## Acceptable deviations

- Drop the KPI row if there are genuinely no numeric facts (rare — even "last updated" works).
- Sticky TOC sidebar on viewports >1080px when the doc has 8+ sections.
- A small sparkline or chart in a recurring report — an eye-landing point for skimmers.
- Two-column layout for sections where a diagram pairs tightly with its caption.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall