メニュー
Render a self-contained HTML report or dashboard (task summaries, code/operational audits) in one canonical dark-terminal shell. NOT for markdown-only output or the mermaid viewer (use mermaid-viewer).
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/atretyak1985/swarmery --skill html-reportingコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Create and drive a /board card as a mirror of a jira-task-runner run: born in triage, moved straight to in_progress, finished at in_review or done. The card MUST NEVER pass through the todo column -- see 'The load-bearing rule' below for why. Covers projectId resolution, the exact POST/PATCH bodies, idempotent reuse via the jira-ticket label, the daemon-unavailable fallback, and dry-run request-body printing. NOT for talking to Jira itself (that's jira-tasks / the Atlassian MCP tools) and NOT for working-repo/config resolution (that's jira-config).
Document an epic/feature architecture with the C4 model -- system context/container/component/dynamic diagrams as Mermaid .mmd plus a narrative doc. NOT for rendering an existing .mmd (use mermaid-viewer).
Generate or refresh the repo-wide architecture map — architecture-out/architecture-map.json (machine contract with named flows) + architecture-map.html (self-contained viewer). Use when the user asks for an architecture map, repo map, "/architecture-map", or when an agent needs a fresh machine-readable architecture overview. NOT for per-epic C4 deep-dives (use c4-architecture-docs) and NOT for building the knowledge graph itself (use /graphify).
SOC 職業分類に基づく
SKILL.md を表示中
| name | html-reporting |
| version | 1.0.0 |
| owner | swarmery-core |
| description | Render a self-contained HTML report or dashboard (task summaries, code/operational audits) in one canonical dark-terminal shell. NOT for markdown-only output or the mermaid viewer (use mermaid-viewer). |
| disable-model-invocation | true |
| color | teal |
| docs | {"status":"reviewed","source_sha":"70bbc7ba9aea","updated":"2026-08-06T00:00:00.000Z"} |
Provide a single, reusable, self-contained HTML shell (inline CSS, no external assets, no JS required) so that every report an agent emits — @summary-generator's Phase 8 dashboard and @code-auditor's Phase 5 audit — shares one visual language: a dark "terminal" theme with severity-coded cards, metric tables, and collapsible sections.
This skill produces presentation only. It does not gather data, measure metrics, or run analysis — the calling agent supplies the content and this skill wraps it.
phases/*.html or SUMMARY-adjacent path)@summary-generator renders the optional Phase 8 HTML dashboard@code-auditor renders the Phase 5 audit report (health score + P0–P3 backlog)summary-templates Step 4)SUMMARY.md / 05-audit.md stay markdown; HTML is the optional mirrormermaid-viewer.html document onlysummary-templates (summaries) or the audit agent's findings; this skill only wraps itNo tooling required. Output is a single self-contained .html file (inline <style>, no CDN, no build step). It opens directly in any browser and survives being copied between machines.
| Input | Required | Description |
|---|---|---|
| Report kind | Yes | summary (Phase 8) or audit (Phase 5) — selects the section skeleton |
| Title | Yes | Report <h1> text |
| Sections | Yes | The already-authored content blocks (metrics, findings, backlog, etc.) |
| Output path | Yes | Where to write, e.g. …/{slug}/phases/05-audit.html or phases/08-summary.html |
| Health score | Audit only | Integer 1–10 for the audit header badge |
Format: one self-contained HTML file at the caller's output path.
Length budget: body content ≤ 300 lines (summary) / ≤ 500 lines (audit). Consolidate similar findings to stay within budget; never pad. The shell CSS does not count against the content budget.
Copy this shell verbatim, then fill {{TITLE}} and the <main> body. Severity classes: .sev-p0/.sev-p1/.sev-p2/.sev-p3 for audit tiers; .ok/.warn/.crit badges for status.
<!DOCTYPE html><html lang="uk"><head><meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>{{TITLE}}</title>
<style>
:root{--bg:#0d1117;--panel:#161b22;--border:#30363d;--text:#e6edf3;
--muted:#8b949e;--accent:#58a6ff;--mono:"SF Mono","JetBrains Mono",Menlo,monospace;
--p0:#f85149;--p1:#d29922;--p2:#58a6ff;--p3:#8b949e;
--ok:#3fb950;--warn:#d29922;--crit:#f85149;}
*{box-sizing:border-box}
body{margin:0;background:(--bg);:(--text);:;
:(--mono);-webkit-:antialiased}
{:;: auto;: }
{: solid (--border);:;:}
{:;: .} {:;: .}
{:(--muted);:.}
{:inline-block;:;:}
{:(--panel);: solid (--border);:;
: ;:. }
{: solid (--p0)} {: solid (--p1)}
{: solid (--p2)} {: solid (--p3)}
{:;:collapse;: ;:.}
,{;:. .;: solid (--border)}
{:(--muted);:uppercase;:.;:.}
{:;:. .;:;:.}
{:inline-block;:. .;:;:.;:}
{:(,,,.);:(--ok)}
{:(,,,.);:(--warn)}
{:(,,,.);:(--crit)}
{:(--panel);: solid (--border);:;:. ;:. }
{:pointer;:}
{:(--accent)}
{{TITLE}}{{META}}
{{BODY}}
summary → sections: Status header → Metrics table → per-role <details> "How to Use" → Next steps <ul> → Known issues .card.crit. Mirror the markdown produced by summary-templates.audit → sections: Executive summary (health .score /10) → Metrics table (current vs target) → Dimension Coverage table → P0–P3 backlog as .card.sev-pN blocks, each with What → Risk/Cost → Fix → How-to-verify → Engineering Standards.Checkpoint: skeleton chosen for the report kind.
Paste the shell, set {{TITLE}}/{{META}}, drop authored content into <main>. Use the severity classes; never invent inline colors.
Checkpoint: all placeholders replaced; no {{…}} left.
Status header → <h1> + .badge; metrics → <table>; per-role guidance → <details>; findings → .card.sev-pN; positives → .badge.ok.
Checkpoint: every section uses a shell component, not ad-hoc HTML.
No external src/href to CDNs; no <script> unless explicitly needed. Count <main> lines against the budget (300 summary / 500 audit) and trim lowest-priority sections first.
Checkpoint: file opens offline; within budget.
Write to the caller's output path. Confirm the file exists and is non-empty (test -s).
Checkpoint: artifact on disk.
.html (no CDN/external assets){{…}} placeholders replaced.sev-p0..p3 / .badge ok|warn|crit classes, not ad-hoc colors<main> within length budget (300 summary / 500 audit)test -s)This skill turns content you have already written into a single self-contained HTML file — one dark terminal-style page with inline CSS, no external assets, and no build step. Every report an agent emits shares the same visual language: severity-coded cards, metric tables, and collapsible sections. It formats what you hand it; it never gathers data, computes metrics, or invents a number.
SUMMARY.md or audit markdown is the source of truth; HTML is the optional mirror.mermaid-viewer instead..html document.Skill(skill: "core:html-reporting")
Invoke it once the section content exists, and pass the report kind, title, sections, and output path along with the call.
summary or audit; selects the section skeleton — required.<h1> — required..html file is written — required.One .html file at your output path, written and verified non-empty on disk. It opens offline in any browser and survives being copied between machines. The <main> body stays within a budget of 300 lines for a summary and 500 for an audit; lowest-priority sections are trimmed rather than padded.
Skill(skill: "core:html-reporting")
kind: audit
title: "Audit — orders/line-items"
output: .../{slug}/phases/05-audit.html
health score: 6
sections: metrics table, dimension coverage, 9 findings across P0–P3
The skill picks the audit skeleton, pastes the canonical shell, renders the score badge and tables, and emits each finding as a .card.sev-pN block with What → Risk/Cost → Fix → How-to-verify. You end up with one offline HTML file at that path.