[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7FD9158Z1WM5MX95SGVH9VQ":3,"skill-same:01M7FD9158Z1WM5MX95SGVH9VQ":62},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7FD9158Z1WM5MX95SGVH9VQ","render-html",null,"Render an ARIS Markdown \u002F JSON artifact (IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT, PAPER_PLAN, research-wiki state, etc.) into a single-file HTML view designed for human reading. Use when the user says \"渲染 HTML\", \"出一份 HTML 报告\", \"render html\", \"make this readable\", \"export to html\", or wants a polished web-rendered view of a Markdown artifact.","CZX-LBJ\u002Fdeepseek-science","CZX-LBJ","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F204594689?v=4","01M7FC59WQ9PFXT04MKGJDCFFT","czx","01M7FC59XB2GW0BKWN9JTDBQS7",1,"MIT","https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Frender-html\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: render-html\ndescription: \"Render an ARIS Markdown \u002F JSON artifact (IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT, PAPER_PLAN, research-wiki state, etc.) into a single-file HTML view designed for human reading. Use when the user says \\\"渲染 HTML\\\", \\\"出一份 HTML 报告\\\", \\\"render html\\\", \\\"make this readable\\\", \\\"export to html\\\", or wants a polished web-rendered view of a Markdown artifact.\"\nargument-hint: \"\u003Cinput.md> [--template academic|dashboard] [--out \u003Cpath>] [--title ...] [--state \u003Cstate.json>] [--json \u003Csidecar.json>] [--offline] [--review|--no-review]\"\nallowed-tools: Bash(*), Read, Write, mcp__codex__codex\n---\n\n# \u002Frender-html: Markdown → single-file HTML for human reading\n\n> **Markdown is for writers. HTML is for readers.** ARIS workflow nodes write Markdown (canonical, audit-trail-friendly, machine-parseable). `\u002Frender-html` turns *selected* artifacts into a polished single-file HTML view for the human who actually has to read them. The Markdown stays the source of truth.\n\n## When to use this skill\n\n**Use `\u002Frender-html` for** ARIS artifacts that have a real human reader:\n\n| Artifact | Why HTML helps | Template |\n|----------|----------------|----------|\n| `idea-stage\u002FIDEA_REPORT.md` | Ranked ideas + pilot signal + scores feel like a decision dashboard, not a flat list | `academic` |\n| `review-stage\u002FAUTO_REVIEW.md` (+ `REVIEW_STATE.json`) | Round-by-round score progression + weakness status; pass `--state` to embed the JSON | `academic` |\n| `paper\u002FKILL_ARGUMENT.md` (+ `KILL_ARGUMENT.json`) | Per-point attack\u002Fdefense with `\u003Cdetails>` Q&A cards + red\u002Fyellow\u002Fgreen callouts | `academic` |\n| `research-wiki\u002FSUMMARY.md` (or `research-wiki\u002Findex.md`) | Cross-entity cockpit: papers \u002F ideas \u002F experiments \u002F claims at a glance | `dashboard` |\n| `PAPER_PLAN.md` (optional) | Claims-evidence matrix renders better as a polished table than raw MD | `academic` |\n| `RESUBMIT_REPORT.{md,json}` (optional) | 7-state failure-mode ledger | `academic` or `dashboard` |\n\n**Do NOT use** for:\n- LaTeX paper output — the final reader-facing artifact is PDF, not HTML.\n- `SKILL.md` files — those are internal LLM-facing protocol.\n- `.aris\u002Ftraces\u002F*` review traces — forensic debug, not human display.\n- Every Markdown file in your project — only artifacts that benefit from sticky TOC, callouts, math, or score progressions.\n\n## Core invariants\n\n- **MD \u002F JSON is canonical, HTML is generated view.** Edit the source, then re-render. Do not hand-edit the HTML.\n- **Cross-model review at the artifact boundary** (ARIS invariant). Academic-template HTML — used for the artifacts humans actually read (IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT, PAPER_PLAN) — is reviewed by a fresh cross-family Codex thread before being claimed as a finished view. Dashboard-template HTML (cockpit \u002F debug views) skips review by default but accepts `--review` to force it. See § *HTML Review Gate* below.\n- **Drift detection.** Every rendered HTML embeds the source path, SHA256, and generation timestamp in `\u003Cmeta>` tags AND in the visible page header. If the HTML and source diverge, the meta tells you which version of the source produced it.\n- **Single-file output.** No build system, no separate CSS, no `node_modules`. Just one `.html`.\n- **CDN-friendly default, `--offline` fallback.** MathJax 3 and highlight.js load from `cdn.jsdelivr.net` by default. Pass `--offline` to skip both — math will appear as raw `$x$`, code blocks won't get syntax highlighting, but everything stays readable.\n- **Pure stdlib helper.** `render_html.py` uses only `re`, `html`, `hashlib`, `json`, `datetime`, `pathlib`, `argparse`, `sys`. No pip install required.\n- **Defense-in-depth XSS sanitization.** The helper strips `\u003Cscript>`\u002F`\u003Cstyle>`\u002F`\u003Ciframe>`\u002F`\u003Cobject>`\u002F`\u003Cembed>`\u002F`\u003Cform>`\u002F`\u003Cinput>`\u002F`\u003Cbutton>`\u002F`\u003Clink>`\u002F`\u003Cmeta>`\u002F`\u003Cbase>` tags, all `on*` event-handler attributes (`onclick`, `onload`, …), and rewrites `javascript:`\u002F`vbscript:`\u002F`data:` href\u002Fsrc\u002Faction schemes to `#blocked-unsafe-url:`. ARIS workflow artifacts should not contain these in the first place, but the sanitizer is the safety net in case an LLM hallucinates one. Markdown text content is HTML-escaped separately and never reaches the sanitizer.\n\n## Tool Location\n\nArch C self-contained: the canonical implementation lives at `skills\u002Frender-html\u002Fscripts\u002Frender_html.py` (this SKILL's own `scripts\u002F` subdirectory), together with its templates at `skills\u002Frender-html\u002Fscripts\u002Ftemplates\u002F{academic,dashboard}.html`. The helper is new — no legacy `tools\u002F` shim exists.\n\nResolve `$RENDER_HTML` with the hybrid chain (Layer 0 prefers the self-contained location for the owning SKILL; Layers 1-3 are the shared-runtime chain documented in [`shared-references\u002Fintegration-contract.md`](..\u002Fshared-references\u002Fintegration-contract.md) §2, **Policy A — skill-local gate**):\n\n```bash\n# Layer 0: self-contained (CC 1.0+ exposes $CLAUDE_SKILL_DIR).\nRENDER_HTML=\"\"\nif [ -n \"${CLAUDE_SKILL_DIR:-}\" ] && [ -f \"$CLAUDE_SKILL_DIR\u002Fscripts\u002Frender_html.py\" ]; then\n  RENDER_HTML=\"$CLAUDE_SKILL_DIR\u002Fscripts\u002Frender_html.py\"\nfi\n# Layers 1-3: shared-runtime chain (non-CC hosts + manual installs).\nif [ -z \"$RENDER_HTML\" ]; then\n  cd \"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || pwd)\" || exit 1\n  if [ -z \"${ARIS_REPO:-}\" ] && [ -f .aris\u002Finstalled-skills.txt ]; then\n      ARIS_REPO=$(awk -F'\\t' '$1==\"repo_root\"{print $2; exit}' .aris\u002Finstalled-skills.txt 2>\u002Fdev\u002Fnull) || true\n  fi\n  RENDER_HTML=\".aris\u002Fskills\u002Frender-html\u002Fscripts\u002Frender_html.py\"\n  [ -f \"$RENDER_HTML\" ] || RENDER_HTML=\"skills\u002Frender-html\u002Fscripts\u002Frender_html.py\"\n  [ -f \"$RENDER_HTML\" ] || { [ -n \"${ARIS_REPO:-}\" ] && RENDER_HTML=\"$ARIS_REPO\u002Fskills\u002Frender-html\u002Fscripts\u002Frender_html.py\"; }\n  [ -f \"$RENDER_HTML\" ] || RENDER_HTML=\"\"\nfi\n[ -z \"$RENDER_HTML\" ] && {\n  echo \"ERROR: render_html.py not resolved (layer 0: \\$CLAUDE_SKILL_DIR\u002Fscripts\u002F; layers 1-3: .aris\u002Fskills\u002Frender-html\u002Fscripts\u002F, skills\u002Frender-html\u002Fscripts\u002F, \\$ARIS_REPO\u002Fskills\u002Frender-html\u002Fscripts\u002F).\" >&2\n  echo \"       \u002Frender-html cannot produce HTML output. Fix: rerun bash tools\u002Finstall_aris.sh, or copy from \\$ARIS_REPO\u002Fskills\u002Frender-html\u002Fscripts\u002F.\" >&2\n  exit 1\n}\n```\n\n## Invocation\n\n```bash\n# Default: academic template, output to \u003Cinput>.html alongside source\npython3 \"$RENDER_HTML\" idea-stage\u002FIDEA_REPORT.md\n\n# Dashboard template for cockpit-style views\npython3 \"$RENDER_HTML\" research-wiki\u002FSUMMARY.md --template dashboard\n\n# Custom output path + title + eyebrow\npython3 \"$RENDER_HTML\" review-stage\u002FAUTO_REVIEW.md \\\n  --out review-stage\u002FAUTO_REVIEW.html \\\n  --title \"Auto Review — overnight run\" \\\n  --eyebrow \"Workflow 2\"\n\n# Embed sidecar state JSON (rendered as a folded \u003Cdetails> JSON block at end)\npython3 \"$RENDER_HTML\" review-stage\u002FAUTO_REVIEW.md \\\n  --state review-stage\u002FREVIEW_STATE.json\n\n# Embed sidecar JSON (e.g., for KILL_ARGUMENT.md + KILL_ARGUMENT.json)\npython3 \"$RENDER_HTML\" paper\u002FKILL_ARGUMENT.md \\\n  --json paper\u002FKILL_ARGUMENT.json \\\n  --title \"Kill Argument — adversarial review\"\n\n# Offline (no CDN; math + code render as plain text)\npython3 \"$RENDER_HTML\" idea-stage\u002FIDEA_REPORT.md --offline\n\n# JSON-only input (wrapped in a \u003Cpre>\u003Ccode class=\"language-json\"> block)\npython3 \"$RENDER_HTML\" review-stage\u002FREVIEW_STATE.json --template dashboard\n\n# Language attr (default zh-CN; set for English-primary artifacts)\npython3 \"$RENDER_HTML\" docs\u002FSKILLS_CATALOG.md --lang en\n\n# Skip review (academic template otherwise reviews by default)\npython3 \"$RENDER_HTML\" idea-stage\u002FIDEA_REPORT.md\n# … then the skill skips the mcp__codex__codex review step if --no-review\n# was on the command line. Pass --review to a dashboard render to force it.\n```\n\nThe `--review` \u002F `--no-review` flags are parsed by the SKILL orchestrator\n(Claude Code), not by `render_html.py`. The helper itself stays pure\nstdlib and never calls MCP. See § *HTML Review Gate* below for the\nexact resolution and prompt.\n\n## Workflow\n\n### Step 1: Identify the artifact\n\nFrom `$ARGUMENTS`, determine what to render. Common patterns:\n\n- \"render IDEA_REPORT\" → `idea-stage\u002FIDEA_REPORT.md`, academic template\n- \"make AUTO_REVIEW readable\" → `review-stage\u002FAUTO_REVIEW.md` + `--state review-stage\u002FREVIEW_STATE.json`\n- \"show the kill argument as HTML\" → `paper\u002FKILL_ARGUMENT.md` + `--json paper\u002FKILL_ARGUMENT.json`\n- \"research-wiki dashboard\" → look for `research-wiki\u002FSUMMARY.md` or generate from raw entity counts (Phase 2 work; for Phase 1 just render `SUMMARY.md` if present, else fall back to listing top-level wiki structure)\n\n### Step 2: Pick template\n\n- `academic` (default): linear long-form. Sticky TOC sidebar + serif body + callouts + tables + math + Q&A `\u003Cdetails>`. Use for IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT, PAPER_PLAN.\n- `dashboard`: grid layout, smaller font, denser metrics cards, no TOC sidebar. Use for research-wiki cockpit, RESUBMIT_REPORT, multi-project overviews.\n\n### Step 3: Run the helper\n\nUse the resolver above to get `$RENDER_HTML`, then invoke. The script writes the HTML alongside the source by default (or wherever `--out` says) and prints a one-line confirmation including the source SHA256 prefix.\n\n### Step 4: HTML Review Gate (cross-model)\n\n**Decide whether to run review.** Per ARIS invariant \"executor must not judge its own output\", the academic-template HTML is reviewed by a fresh cross-family Codex thread before being claimed as a delivered view. Resolution:\n\n```\nshould_review = explicit --review present\n             or (template == \"academic\" and --no-review NOT present)\n```\n\nSo:\n\n- `--template academic` (default) → **review by default**. Skip with `--no-review`.\n- `--template dashboard` → no review by default. Force with `--review`.\n- Phase 2 workflow auto-emit (activated 2026-05) selects per-skill via the RENDER_HTML hooks documented below — interim views default to `--no-review`, final \u002F audit-class deliverables default to full gate.\n\n**If `should_review` is true**, fire a fresh `mcp__codex__codex` thread (NEVER `codex-reply`; pin `model: gpt-6-astra` + `config: {\"model_reasoning_effort\": \"xhigh\"}` per `..\u002Fshared-references\u002Freviewer-routing.md`) with the prompt below. The reviewer reads the source MD + generated HTML directly; it does **not** see this skill's intermediate state.\n\n**Scope of review (narrow on purpose).** The HTML reviewer audits **render fidelity \u002F safety \u002F structure only** — not claim truthfulness. Claim audit belongs upstream (`\u002Fpaper-claim-audit`, `\u002Fresearch-review`, `\u002Fresult-to-claim`). Specifically the reviewer checks:\n\n1. **Information fidelity** — every section, claim, table, code block, list, math snippet, sidecar payload is present in the HTML (no silent drop)\n2. **Structural integrity** — heading hierarchy \u002F nested lists \u002F table cells \u002F code fences \u002F `\u003Cdetails>` blocks \u002F math delimiters survive parsing\n3. **Callout routing** — `> 🚨 …` got `.callout-bad`, `> 💡 …` got `.callout-info`, etc.\n4. **Safety \u002F escaping** — no raw `\u003Cscript>` \u002F `onclick=` \u002F `javascript:` \u002F unreplaced placeholders \u002F template-leak survives\n5. **Expected-difference allowance** — frontmatter strip, generated header\u002Ffooter\u002Fmeta, TOC insert, sanitized unsafe HTML are all expected, not flagged\n\n**Codex prompt (mandatory shape).** Send this as a fresh thread (`mcp__codex__codex`, NOT `codex-reply`):\n\n```\nYou are an independent ARIS HTML render auditor. This is a fresh review thread.\n\nRead these files directly:\n- Source artifact: \u003CABS path to source.md or source.json>\n- Generated HTML:  \u003CABS path to out.html>\n- Optional sidecars: \u003Cstate.json>, \u003Ckill_argument.json> (if any)\n\nTask: Audit whether the generated HTML is a faithful, safe, structurally\nusable view of the source artifact. Do NOT judge whether the research\nclaims are true. Judge only rendering fidelity.\n\nChecks:\n1. Information fidelity — sections \u002F claims \u002F tables \u002F code blocks \u002F\n   lists \u002F math \u002F sidecars not silently dropped or materially altered.\n2. Structural integrity — heading hierarchy, tables, nested lists,\n   code fences, details\u002Fsummary, math delimiters preserved.\n3. Callout routing — warning\u002Fcritical\u002Fgood\u002Finfo blockquotes map to\n   appropriate CSS classes when present.\n4. Safety\u002Fescaping — no unexpected raw script\u002Fstyle\u002Fiframe\u002Fform\u002F\n   event-handler\u002Fjavascript\u002Fdata URL survives from source.\n5. Placeholder\u002Ftemplate leakage — no unreplaced {{PLACEHOLDER}} or\n   parser private placeholder character appears.\n6. Expected differences — frontmatter strip, generated header\u002Ffooter\u002F\n   meta, TOC insertion, sanitized unsafe HTML are EXPECTED and not a\n   defect.\n\nReturn STRICT JSON first, then a short prose note:\n\n{\n  \"verdict\": \"PASS|WARN|FAIL|ERROR\",\n  \"checks\": {\n    \"source_hash_match\": \"pass|warn|fail|unknown\",\n    \"information_fidelity\": \"pass|warn|fail\",\n    \"structure\": \"pass|warn|fail\",\n    \"math_code_tables\": \"pass|warn|fail\",\n    \"callouts\": \"pass|warn|fail|not_applicable\",\n    \"safety_escaping\": \"pass|warn|fail\",\n    \"placeholder_leak\": \"pass|warn|fail\"\n  },\n  \"blocking_issues\": [\n    {\"severity\": \"fail\",\n     \"source_location\": \"L\u003Cn>...\",\n     \"html_location\": \"\u003Cselector or near-text>\",\n     \"issue\": \"...\",\n     \"suggested_fix\": \"...\"}\n  ],\n  \"warnings\": [\n    {\"severity\": \"warn\", \"issue\": \"...\", \"suggested_fix\": \"...\"}\n  ],\n  \"summary\": \"one paragraph\"\n}\n\nVerdict rules:\n- PASS: no material fidelity\u002Fsafety issue.\n- WARN: readable output with minor unsupported-Markdown or cosmetic\n  degradation only.\n- FAIL: missing\u002Faltered meaningful content, broken tables\u002Fmath\u002Fcode\u002F\n  callouts that change interpretation, unsafe executable HTML, source\n  hash mismatch, or placeholder leakage.\n- ERROR: files could not be read or audit could not complete.\n```\n\n**Save outputs**:\n\n1. Write the JSON verdict to `\u003Cout_path>.review.json` (sibling to the HTML).\n2. Save the raw codex trace to `.aris\u002Ftraces\u002Frender-html\u002F\u003CYYYY-MM-DD>_run\u003CNN>\u002Freview.{txt,json}` per `shared-references\u002Freview-tracing.md`.\n3. Print a one-line summary to the user: `verdict, N blocking, N warnings, trace: \u003Cpath>`.\n\n**If `verdict == FAIL`**: the HTML is **NOT** a delivered review-passed view. Tell the user the blocking issues, point them at the source (fix MD or template, not the HTML), and re-render. Do not silently overwrite or mark as complete.\n\n**If `verdict == WARN`**: deliver the HTML but surface the warning list. User decides whether to fix or accept.\n\n**If `mcp__codex__codex` is not available** (e.g., user runs `\u002Frender-html` on a Codex-CLI-only setup where Codex MCP isn't wired): emit `verdict: REVIEW_UNAVAILABLE` to the sidecar, do not fabricate `PASS`, and tell the user the HTML was generated but **not** independently reviewed. The user can manually invoke `\u002Fresearch-review` on the source MD or re-run with Codex MCP available.\n\n### Step 5: (Optional) Verify in browser\n\n`open \u003Cout.html>` on macOS, `xdg-open` on Linux, `start` on Windows. Math + code highlighting need internet (CDN); offline mode degrades gracefully to readable text.\n\n## What the helper supports\n\n**Markdown subset** (chosen to match what ARIS workflows actually emit):\n\n- Headings `#`\u002F`##`\u002F`###`\u002F`####` with auto-generated IDs for TOC\n- Paragraphs, bold `**x**`, italic `*x*` \u002F `_x_`, inline code `` `x` ``, strikethrough `~~x~~`, links `[t](url)`, images `![a](url)`\n- Unordered\u002Fordered lists with 2-space nested indentation\n- Code blocks with optional language (` ```python `) — gets `\u003Cpre>\u003Ccode class=\"language-python\">` for highlight.js\n- ASCII-art code blocks (heuristic: many box-drawing chars) → `\u003Cpre class=\"diagram\">` with a distinct cream-yellow background\n- Tables with `:---` \u002F `:---:` \u002F `---:` alignment\n- Blockquotes `> ...` — emoji-prefix detection routes to callout variants:\n  - `⚠️` → `.callout-warn` (Warning)\n  - `💡` \u002F `📝` → `.callout-info` (Tip \u002F Note)\n  - `✅` \u002F `🔒` → `.callout-good` (OK \u002F Guarantee)\n  - `❌` \u002F `🚨` → `.callout-bad` (Blocked \u002F Critical)\n- HTML passthrough for `\u003Cdetails>`, `\u003Csummary>`, `\u003Cdiv>`, `\u003Cfigure>`, `\u003Ctable>`, `\u003Csection>`, etc. (block-level) — used for the existing Round-1\u002F2\u002F3 fix details in `KILL_ARGUMENT.md`\n- LaTeX math: inline `$x$` and display `$$x$$` pass through verbatim to MathJax\n\n**What the helper does NOT support** (intentional simplifications):\n\n- Footnotes (`[^1]`)\n- Definition lists\n- Reference-style links `[label][ref]`\n- Setext-style headings (`===` \u002F `---` underline form — ARIS doesn't emit these)\n\n**Frontmatter (`--- ... ---`) at the very top of the file is stripped before rendering** (SKILL.md frontmatter style); the body that follows is what gets converted.\n\n## Two-phase integration plan (Phase 1 was this skill; Phase 2 is workflow hooks, **activated 2026-05**)\n\n**Phase 1 (originally shipped):** Opt-in only. Users explicitly called `\u002Frender-html \u003Cartifact.md>` after a workflow completes.\n\n**Phase 2 (activated 2026-05):** Eight key skills now auto-emit HTML at workflow termination, each guarded by a `RENDER_HTML = true` constant that the user can flip off per skill (or pass `— render html: false` at invocation).\n\n**Cost-tiered review gate** (per artifact type):\n\n| Skill | Artifact | Review tier | Why |\n|---|---|---|---|\n| `\u002Fidea-discovery` | `idea-stage\u002FIDEA_REPORT.md` | `--no-review` | Interim; source went through novelty + cross-model review already |\n| `\u002Fauto-review-loop` | `review-stage\u002FAUTO_REVIEW.md` | `--no-review` | Per-round log; the loop itself is the cross-model review |\n| `\u002Fresearch-pipeline` | `NARRATIVE_REPORT.md` | `--no-review` | Internal Stage 5 handoff to `\u002Fpaper-writing`; not reviewer-facing — upstream Stage 4 auto-review-loop already gated the claims |\n| `\u002Fkill-argument` | `\u003Cpaper-dir>\u002FKILL_ARGUMENT.md` | full gate | Audit-class artifact; matches skill's own cross-model invariant |\n| `\u002Fpaper-claim-audit` | `paper\u002FPAPER_CLAIM_AUDIT.md` | full gate | Audit-class artifact |\n| `\u002Fcitation-audit` | `paper\u002FCITATION_AUDIT.md` | full gate | Audit-class artifact |\n| `\u002Fproof-checker` | `PROOF_AUDIT.md` | full gate | Audit-class artifact |\n| `\u002Frebuttal` | `REBUTTAL_DRAFT_rich.md` | full gate | Pre-submission deliverable |\n| `\u002Fpaper-writing` | (no auto-HTML) | n\u002Fa | The final reader artifact is PDF |\n| `\u002Fresearch-wiki` | `research-wiki\u002Findex.html` dashboard | future | Phase 2.1 — dashboard template work, not in this round |\n\n`--no-review` for interim reports is intentional: the source MD has already been reviewed by the producing skill's own cross-model gate, and the HTML render only converts structure — it doesn't add new claims. Full review at every checkpoint would multiply Codex calls per pipeline by 4-6×.\n\nTo disable HTML output for a specific skill: set `RENDER_HTML = false` in the skill's constants block or pass `— render html: false`. To globally disable: skip with environment variable `ARIS_RENDER_HTML=0` (planned, not yet implemented; flag per skill in the meantime).\n\n## Customizing the templates\n\nThe two templates live at `skills\u002Frender-html\u002Fscripts\u002Ftemplates\u002F{academic,dashboard}.html`. Each is a single self-contained HTML file with inline CSS using `{{PLACEHOLDER}}` substitution. To customize:\n\n1. Copy one of the templates to a new name, e.g., `my_brand.html`.\n2. Edit the CSS variables in `:root { ... }` to change colors, the font stack, or layout dimensions.\n3. Add the template name to the `--template` choices in `render_html.py` `argparse`.\n4. Re-run `\u002Frender-html \u003Cinput> --template my_brand`.\n\nThe default templates are derived from the user's own academic-newspaper tutorial style (Source Serif Pro + Songti SC, 3-color palette, sticky TOC, low-flash). Stay close to that idiom for ARIS artifacts unless you have a specific reason to break the visual language.\n\n## External alternatives (for richer surfaces)\n\nFor deck \u002F poster \u002F Xiaohongshu card \u002F tweet card \u002F data report style outputs, point users to **[html-anything](https:\u002F\u002Fgithub.com\u002Fnexu-io\u002Fhtml-anything)** (Apache-2.0, 3000⭐ at time of writing). It ships 75 SKILL.md templates across 9 surfaces and detects 8 coding-agent CLIs (including Claude Code, Codex, Copilot). ARIS does *not* depend on it — `\u002Frender-html` covers ARIS-native artifacts; html-anything is the recommended path for richer publishing surfaces.\n\n## Key rules\n\n- **Do not auto-render every Markdown file.** Only artifacts on the whitelist above. File proliferation is the main anti-pattern.\n- **Do not hand-edit the generated HTML.** Edit the source, then re-render. The embedded SHA256 in the HTML meta tells you if the source has changed since render.\n- **academic-template HTML is a reviewed artifact by default**, not raw output. Cross-model Codex review (fresh thread) gates the academic deliverables — the same way `\u002Fproof-checker`, `\u002Fpaper-claim-audit`, `\u002Fcitation-audit`, `\u002Fkill-argument` gate their respective products. `--no-review` is appropriate for **interim auto-emits** (e.g., `idea-stage\u002FIDEA_REPORT.html`, per-round `review-stage\u002FAUTO_REVIEW.html`) where the source MD has already passed an upstream cross-model gate — the HTML render is then a structural conversion, not a new claim audit. For **shipped \u002F reviewer-facing \u002F audit-class** outputs, keep the full gate.\n- **The reviewer audits rendering, not research.** Claim truthfulness is owned upstream by `\u002Fpaper-claim-audit`, `\u002Fresult-to-claim`, `\u002Fresearch-review`. The HTML reviewer asks: \"did the renderer faithfully + safely convert this source?\" — nothing more.\n- **CDN dependency is opt-out, not opt-in.** Most users have internet; `--offline` is for air-gapped runs \u002F archival.\n- **The default style is academic-newspaper, not marketing-flashy.** Match the existing ARIS tonal voice. If you want decks\u002Fposters\u002Fsocial cards, point users to html-anything.\n- **Pure stdlib only.** Adding a `pip install` dependency to `render_html.py` requires an explicit decision — the helper currently has none. MCP calls live in the skill orchestrator, never in the helper script.\n","# \u002Frender-html: Markdown → single-file HTML for human reading\n\n> **Markdown is for writers. HTML is for readers.** ARIS workflow nodes write Markdown (canonical, audit-trail-friendly, machine-parseable). `\u002Frender-html` turns *selected* artifacts into a polished single-file HTML view for the human who actually has to read them. The Markdown stays the source of truth.\n\n## When to use this skill\n\n**Use `\u002Frender-html` for** ARIS artifacts that have a real human reader:\n\n| Artifact | Why HTML helps | Template |\n|----------|----------------|----------|\n| `idea-stage\u002FIDEA_REPORT.md` | Ranked ideas + pilot signal + scores feel like a decision dashboard, not a flat list | `academic` |\n| `review-stage\u002FAUTO_REVIEW.md` (+ `REVIEW_STATE.json`) | Round-by-round score progression + weakness status; pass `--state` to embed the JSON | `academic` |\n| `paper\u002FKILL_ARGUMENT.md` (+ `KILL_ARGUMENT.json`) | Per-point attack\u002Fdefense with `\u003Cdetails>` Q&A cards + red\u002Fyellow\u002Fgreen callouts | `academic` |\n| `research-wiki\u002FSUMMARY.md` (or `research-wiki\u002Findex.md`) | Cross-entity cockpit: papers \u002F ideas \u002F experiments \u002F claims at a glance | `dashboard` |\n| `PAPER_PLAN.md` (optional) | Claims-evidence matrix renders better as a polished table than raw MD | `academic` |\n| `RESUBMIT_REPORT.{md,json}` (optional) | 7-state failure-mode ledger | `academic` or `dashboard` |\n\n**Do NOT use** for:\n- LaTeX paper output — the final reader-facing artifact is PDF, not HTML.\n- `SKILL.md` files — those are internal LLM-facing protocol.\n- `.aris\u002Ftraces\u002F*` review traces — forensic debug, not human display.\n- Every Markdown file in your project — only artifacts that benefit from sticky TOC, callouts, math, or score progressions.\n\n## Core invariants\n\n- **MD \u002F JSON is canonical, HTML is generated view.** Edit the source, then re-render. Do not hand-edit the HTML.\n- **Cross-model review at the artifact boundary** (ARIS invariant). Academic-template HTML — used for the artifacts humans actually read (IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT, PAPER_PLAN) — is reviewed by a fresh cross-family Codex thread before being claimed as a finished view. Dashboard-template HTML (cockpit \u002F debug views) skips review by default but accepts `--review` to force it. See § *HTML Review Gate* below.\n- **Drift detection.** Every rendered HTML embeds the source path, SHA256, and generation timestamp in `\u003Cmeta>` tags AND in the visible page header. If the HTML and source diverge, the meta tells you which version of the source produced it.\n- **Single-file output.** No build system, no separate CSS, no `node_modules`. Just one `.html`.\n- **CDN-friendly default, `--offline` fallback.** MathJax 3 and highlight.js load from `cdn.jsdelivr.net` by default. Pass `--offline` to skip both — math will appear as raw `$x$`, code blocks won't get syntax highlighting, but everything stays readable.\n- **Pure stdlib helper.** `render_html.py` uses only `re`, `html`, `hashlib`, `json`, `datetime`, `pathlib`, `argparse`, `sys`. No pip install required.\n- **Defense-in-depth XSS sanitization.** The helper strips `\u003Cscript>`\u002F`\u003Cstyle>`\u002F`\u003Ciframe>`\u002F`\u003Cobject>`\u002F`\u003Cembed>`\u002F`\u003Cform>`\u002F`\u003Cinput>`\u002F`\u003Cbutton>`\u002F`\u003Clink>`\u002F`\u003Cmeta>`\u002F`\u003Cbase>` tags, all `on*` event-handler attributes (`onclick`, `onload`, …), and rewrites `javascript:`\u002F`vbscript:`\u002F`data:` href\u002Fsrc\u002Faction schemes to `#blocked-unsafe-url:`. ARIS workflow artifacts should not contain these in the first place, but the sanitizer is the safety net in case an LLM hallucinates one. Markdown text content is HTML-escaped separately and never reaches the sanitizer.\n\n## Tool Location\n\nArch C self-contained: the canonical implementation lives at `skills\u002Frender-html\u002Fscripts\u002Frender_html.py` (this SKILL's own `scripts\u002F` subdirectory), together with its templates at `skills\u002Frender-html\u002Fscripts\u002Ftemplates\u002F{academic,dashboard}.html`. The helper is new — no legacy `tools\u002F` shim exists.\n\nResolve `$RENDER_HTML` with the hybrid chain (Layer 0 prefers the self-contained location for the owning SKILL; Layers 1-3 are the shared-runtime chain documented in [`shared-references\u002Fintegration-contract.md`](..\u002Fshared-references\u002Fintegration-contract.md) §2, **Policy A — skill-local gate**):\n\n```bash\n# Layer 0: self-contained (CC 1.0+ exposes $CLAUDE_SKILL_DIR).\nRENDER_HTML=\"\"\nif [ -n \"${CLAUDE_SKILL_DIR:-}\" ] && [ -f \"$CLAUDE_SKILL_DIR\u002Fscripts\u002Frender_html.py\" ]; then\n  RENDER_HTML=\"$CLAUDE_SKILL_DIR\u002Fscripts\u002Frender_html.py\"\nfi\n# Layers 1-3: shared-runtime chain (non-CC hosts + manual installs).\nif [ -z \"$RENDER_HTML\" ]; then\n  cd \"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || pwd)\" || exit 1\n  if [ -z \"${ARIS_REPO:-}\" ] && [ -f .aris\u002Finstalled-skills.txt ]; then\n      ARIS_REPO=$(awk -F'\\t' '$1==\"repo_root\"{print $2; exit}' .aris\u002Finstalled-skills.txt 2>\u002Fdev\u002Fnull) || true\n  fi\n  RENDER_HTML=\".aris\u002Fskills\u002Frender-html\u002Fscripts\u002Frender_html.py\"\n  [ -f \"$RENDER_HTML\" ] || RENDER_HTML=\"skills\u002Frender-html\u002Fscripts\u002Frender_html.py\"\n  [ -f \"$RENDER_HTML\" ] || { [ -n \"${ARIS_REPO:-}\" ] && RENDER_HTML=\"$ARIS_REPO\u002Fskills\u002Frender-html\u002Fscripts\u002Frender_html.py\"; }\n  [ -f \"$RENDER_HTML\" ] || RENDER_HTML=\"\"\nfi\n[ -z \"$RENDER_HTML\" ] && {\n  echo \"ERROR: render_html.py not resolved (layer 0: \\$CLAUDE_SKILL_DIR\u002Fscripts\u002F; layers 1-3: .aris\u002Fskills\u002Frender-html\u002Fscripts\u002F, skills\u002Frender-html\u002Fscripts\u002F, \\$ARIS_REPO\u002Fskills\u002Frender-html\u002Fscripts\u002F).\" >&2\n  echo \"       \u002Frender-html cannot produce HTML output. Fix: rerun bash tools\u002Finstall_aris.sh, or copy from \\$ARIS_REPO\u002Fskills\u002Frender-html\u002Fscripts\u002F.\" >&2\n  exit 1\n}\n```\n\n## Invocation\n\n```bash\n# Default: academic template, output to \u003Cinput>.html alongside source\npython3 \"$RENDER_HTML\" idea-stage\u002FIDEA_REPORT.md\n\n# Dashboard template for cockpit-style views\npython3 \"$RENDER_HTML\" research-wiki\u002FSUMMARY.md --template dashboard\n\n# Custom output path + title + eyebrow\npython3 \"$RENDER_HTML\" review-stage\u002FAUTO_REVIEW.md \\\n  --out review-stage\u002FAUTO_REVIEW.html \\\n  --title \"Auto Review — overnight run\" \\\n  --eyebrow \"Workflow 2\"\n\n# Embed sidecar state JSON (rendered as a folded \u003Cdetails> JSON block at end)\npython3 \"$RENDER_HTML\" review-stage\u002FAUTO_REVIEW.md \\\n  --state review-stage\u002FREVIEW_STATE.json\n\n# Embed sidecar JSON (e.g., for KILL_ARGUMENT.md + KILL_ARGUMENT.json)\npython3 \"$RENDER_HTML\" paper\u002FKILL_ARGUMENT.md \\\n  --json paper\u002FKILL_ARGUMENT.json \\\n  --title \"Kill Argument — adversarial review\"\n\n# Offline (no CDN; math + code render as plain text)\npython3 \"$RENDER_HTML\" idea-stage\u002FIDEA_REPORT.md --offline\n\n# JSON-only input (wrapped in a \u003Cpre>\u003Ccode class=\"language-json\"> block)\npython3 \"$RENDER_HTML\" review-stage\u002FREVIEW_STATE.json --template dashboard\n\n# Language attr (default zh-CN; set for English-primary artifacts)\npython3 \"$RENDER_HTML\" docs\u002FSKILLS_CATALOG.md --lang en\n\n# Skip review (academic template otherwise reviews by default)\npython3 \"$RENDER_HTML\" idea-stage\u002FIDEA_REPORT.md\n# … then the skill skips the mcp__codex__codex review step if --no-review\n# was on the command line. Pass --review to a dashboard render to force it.\n```\n\nThe `--review` \u002F `--no-review` flags are parsed by the SKILL orchestrator\n(Claude Code), not by `render_html.py`. The helper itself stays pure\nstdlib and never calls MCP. See § *HTML Review Gate* below for the\nexact resolution and prompt.\n\n## Workflow\n\n### Step 1: Identify the artifact\n\nFrom `$ARGUMENTS`, determine what to render. Common patterns:\n\n- \"render IDEA_REPORT\" → `idea-stage\u002FIDEA_REPORT.md`, academic template\n- \"make AUTO_REVIEW readable\" → `review-stage\u002FAUTO_REVIEW.md` + `--state review-stage\u002FREVIEW_STATE.json`\n- \"show the kill argument as HTML\" → `paper\u002FKILL_ARGUMENT.md` + `--json paper\u002FKILL_ARGUMENT.json`\n- \"research-wiki dashboard\" → look for `research-wiki\u002FSUMMARY.md` or generate from raw entity counts (Phase 2 work; for Phase 1 just render `SUMMARY.md` if present, else fall back to listing top-level wiki structure)\n\n### Step 2: Pick template\n\n- `academic` (default): linear long-form. Sticky TOC sidebar + serif body + callouts + tables + math + Q&A `\u003Cdetails>`. Use for IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT, PAPER_PLAN.\n- `dashboard`: grid layout, smaller font, denser metrics cards, no TOC sidebar. Use for research-wiki cockpit, RESUBMIT_REPORT, multi-project overviews.\n\n### Step 3: Run the helper\n\nUse the resolver above to get `$RENDER_HTML`, then invoke. The script writes the HTML alongside the source by default (or wherever `--out` says) and prints a one-line confirmation including the source SHA256 prefix.\n\n### Step 4: HTML Review Gate (cross-model)\n\n**Decide whether to run review.** Per ARIS invariant \"executor must not judge its own output\", the academic-template HTML is reviewed by a fresh cross-family Codex thread before being claimed as a delivered view. Resolution:\n\n```\nshould_review = explicit --review present\n             or (template == \"academic\" and --no-review NOT present)\n```\n\nSo:\n\n- `--template academic` (default) → **review by default**. Skip with `--no-review`.\n- `--template dashboard` → no review by default. Force with `--review`.\n- Phase 2 workflow auto-emit (activated 2026-05) selects per-skill via the RENDER_HTML hooks documented below — interim views default to `--no-review`, final \u002F audit-class deliverables default to full gate.\n\n**If `should_review` is true**, fire a fresh `mcp__codex__codex` thread (NEVER `codex-reply`; pin `model: gpt-6-astra` + `config: {\"model_reasoning_effort\": \"xhigh\"}` per `..\u002Fshared-references\u002Freviewer-routing.md`) with the prompt below. The reviewer reads the source MD + generated HTML directly; it does **not** see this skill's intermediate state.\n\n**Scope of review (narrow on purpose).** The HTML reviewer audits **render fidelity \u002F safety \u002F structure only** — not claim truthfulness. Claim audit belongs upstream (`\u002Fpaper-claim-audit`, `\u002Fresearch-review`, `\u002Fresult-to-claim`). Specifically the reviewer checks:\n\n1. **Information fidelity** — every section, claim, table, code block, list, math snippet, sidecar payload is present in the HTML (no silent drop)\n2. **Structural integrity** — heading hierarchy \u002F nested lists \u002F table cells \u002F code fences \u002F `\u003Cdetails>` blocks \u002F math delimiters survive parsing\n3. **Callout routing** — `> 🚨 …` got `.callout-bad`, `> 💡 …` got `.callout-info`, etc.\n4. **Safety \u002F escaping** — no raw `\u003Cscript>` \u002F `onclick=` \u002F `javascript:` \u002F unreplaced placeholders \u002F template-leak survives\n5. **Expected-difference allowance** — frontmatter strip, generated header\u002Ffooter\u002Fmeta, TOC insert, sanitized unsafe HTML are all expected, not flagged\n\n**Codex prompt (mandatory shape).** Send this as a fresh thread (`mcp__codex__codex`, NOT `codex-reply`):\n\n```\nYou are an independent ARIS HTML render auditor. This is a fresh review thread.\n\nRead these files directly:\n- Source artifact: \u003CABS path to source.md or source.json>\n- Generated HTML:  \u003CABS path to out.html>\n- Optional sidecars: \u003Cstate.json>, \u003Ckill_argument.json> (if any)\n\nTask: Audit whether the generated HTML is a faithful, safe, structurally\nusable view of the source artifact. Do NOT judge whether the research\nclaims are true. Judge only rendering fidelity.\n\nChecks:\n1. Information fidelity — sections \u002F claims \u002F tables \u002F code blocks \u002F\n   lists \u002F math \u002F sidecars not silently dropped or materially altered.\n2. Structural integrity — heading hierarchy, tables, nested lists,\n   code fences, details\u002Fsummary, math delimiters preserved.\n3. Callout routing — warning\u002Fcritical\u002Fgood\u002Finfo blockquotes map to\n   appropriate CSS classes when present.\n4. Safety\u002Fescaping — no unexpected raw script\u002Fstyle\u002Fiframe\u002Fform\u002F\n   event-handler\u002Fjavascript\u002Fdata URL survives from source.\n5. Placeholder\u002Ftemplate leakage — no unreplaced {{PLACEHOLDER}} or\n   parser private placeholder character appears.\n6. Expected differences — frontmatter strip, generated header\u002Ffooter\u002F\n   meta, TOC insertion, sanitized unsafe HTML are EXPECTED and not a\n   defect.\n\nReturn STRICT JSON first, then a short prose note:\n\n{\n  \"verdict\": \"PASS|WARN|FAIL|ERROR\",\n  \"checks\": {\n    \"source_hash_match\": \"pass|warn|fail|unknown\",\n    \"information_fidelity\": \"pass|warn|fail\",\n    \"structure\": \"pass|warn|fail\",\n    \"math_code_tables\": \"pass|warn|fail\",\n    \"callouts\": \"pass|warn|fail|not_applicable\",\n    \"safety_escaping\": \"pass|warn|fail\",\n    \"placeholder_leak\": \"pass|warn|fail\"\n  },\n  \"blocking_issues\": [\n    {\"severity\": \"fail\",\n     \"source_location\": \"L\u003Cn>...\",\n     \"html_location\": \"\u003Cselector or near-text>\",\n     \"issue\": \"...\",\n     \"suggested_fix\": \"...\"}\n  ],\n  \"warnings\": [\n    {\"severity\": \"warn\", \"issue\": \"...\", \"suggested_fix\": \"...\"}\n  ],\n  \"summary\": \"one paragraph\"\n}\n\nVerdict rules:\n- PASS: no material fidelity\u002Fsafety issue.\n- WARN: readable output with minor unsupported-Markdown or cosmetic\n  degradation only.\n- FAIL: missing\u002Faltered meaningful content, broken tables\u002Fmath\u002Fcode\u002F\n  callouts that change interpretation, unsafe executable HTML, source\n  hash mismatch, or placeholder leakage.\n- ERROR: files could not be read or audit could not complete.\n```\n\n**Save outputs**:\n\n1. Write the JSON verdict to `\u003Cout_path>.review.json` (sibling to the HTML).\n2. Save the raw codex trace to `.aris\u002Ftraces\u002Frender-html\u002F\u003CYYYY-MM-DD>_run\u003CNN>\u002Freview.{txt,json}` per `shared-references\u002Freview-tracing.md`.\n3. Print a one-line summary to the user: `verdict, N blocking, N warnings, trace: \u003Cpath>`.\n\n**If `verdict == FAIL`**: the HTML is **NOT** a delivered review-passed view. Tell the user the blocking issues, point them at the source (fix MD or template, not the HTML), and re-render. Do not silently overwrite or mark as complete.\n\n**If `verdict == WARN`**: deliver the HTML but surface the warning list. User decides whether to fix or accept.\n\n**If `mcp__codex__codex` is not available** (e.g., user runs `\u002Frender-html` on a Codex-CLI-only setup where Codex MCP isn't wired): emit `verdict: REVIEW_UNAVAILABLE` to the sidecar, do not fabricate `PASS`, and tell the user the HTML was generated but **not** independently reviewed. The user can manually invoke `\u002Fresearch-review` on the source MD or re-run with Codex MCP available.\n\n### Step 5: (Optional) Verify in browser\n\n`open \u003Cout.html>` on macOS, `xdg-open` on Linux, `start` on Windows. Math + code highlighting need internet (CDN); offline mode degrades gracefully to readable text.\n\n## What the helper supports\n\n**Markdown subset** (chosen to match what ARIS workflows actually emit):\n\n- Headings `#`\u002F`##`\u002F`###`\u002F`####` with auto-generated IDs for TOC\n- Paragraphs, bold `**x**`, italic `*x*` \u002F `_x_`, inline code `` `x` ``, strikethrough `~~x~~`, links `[t](url)`, images `![a](url)`\n- Unordered\u002Fordered lists with 2-space nested indentation\n- Code blocks with optional language (` ```python `) — gets `\u003Cpre>\u003Ccode class=\"language-python\">` for highlight.js\n- ASCII-art code blocks (heuristic: many box-drawing chars) → `\u003Cpre class=\"diagram\">` with a distinct cream-yellow background\n- Tables with `:---` \u002F `:---:` \u002F `---:` alignment\n- Blockquotes `> ...` — emoji-prefix detection routes to callout variants:\n  - `⚠️` → `.callout-warn` (Warning)\n  - `💡` \u002F `📝` → `.callout-info` (Tip \u002F Note)\n  - `✅` \u002F `🔒` → `.callout-good` (OK \u002F Guarantee)\n  - `❌` \u002F `🚨` → `.callout-bad` (Blocked \u002F Critical)\n- HTML passthrough for `\u003Cdetails>`, `\u003Csummary>`, `\u003Cdiv>`, `\u003Cfigure>`, `\u003Ctable>`, `\u003Csection>`, etc. (block-level) — used for the existing Round-1\u002F2\u002F3 fix details in `KILL_ARGUMENT.md`\n- LaTeX math: inline `$x$` and display `$$x$$` pass through verbatim to MathJax\n\n**What the helper does NOT support** (intentional simplifications):\n\n- Footnotes (`[^1]`)\n- Definition lists\n- Reference-style links `[label][ref]`\n- Setext-style headings (`===` \u002F `---` underline form — ARIS doesn't emit these)\n\n**Frontmatter (`--- ... ---`) at the very top of the file is stripped before rendering** (SKILL.md frontmatter style); the body that follows is what gets converted.\n\n## Two-phase integration plan (Phase 1 was this skill; Phase 2 is workflow hooks, **activated 2026-05**)\n\n**Phase 1 (originally shipped):** Opt-in only. Users explicitly called `\u002Frender-html \u003Cartifact.md>` after a workflow completes.\n\n**Phase 2 (activated 2026-05):** Eight key skills now auto-emit HTML at workflow termination, each guarded by a `RENDER_HTML = true` constant that the user can flip off per skill (or pass `— render html: false` at invocation).\n\n**Cost-tiered review gate** (per artifact type):\n\n| Skill | Artifact | Review tier | Why |\n|---|---|---|---|\n| `\u002Fidea-discovery` | `idea-stage\u002FIDEA_REPORT.md` | `--no-review` | Interim; source went through novelty + cross-model review already |\n| `\u002Fauto-review-loop` | `review-stage\u002FAUTO_REVIEW.md` | `--no-review` | Per-round log; the loop itself is the cross-model review |\n| `\u002Fresearch-pipeline` | `NARRATIVE_REPORT.md` | `--no-review` | Internal Stage 5 handoff to `\u002Fpaper-writing`; not reviewer-facing — upstream Stage 4 auto-review-loop already gated the claims |\n| `\u002Fkill-argument` | `\u003Cpaper-dir>\u002FKILL_ARGUMENT.md` | full gate | Audit-class artifact; matches skill's own cross-model invariant |\n| `\u002Fpaper-claim-audit` | `paper\u002FPAPER_CLAIM_AUDIT.md` | full gate | Audit-class artifact |\n| `\u002Fcitation-audit` | `paper\u002FCITATION_AUDIT.md` | full gate | Audit-class artifact |\n| `\u002Fproof-checker` | `PROOF_AUDIT.md` | full gate | Audit-class artifact |\n| `\u002Frebuttal` | `REBUTTAL_DRAFT_rich.md` | full gate | Pre-submission deliverable |\n| `\u002Fpaper-writing` | (no auto-HTML) | n\u002Fa | The final reader artifact is PDF |\n| `\u002Fresearch-wiki` | `research-wiki\u002Findex.html` dashboard | future | Phase 2.1 — dashboard template work, not in this round |\n\n`--no-review` for interim reports is intentional: the source MD has already been reviewed by the producing skill's own cross-model gate, and the HTML render only converts structure — it doesn't add new claims. Full review at every checkpoint would multiply Codex calls per pipeline by 4-6×.\n\nTo disable HTML output for a specific skill: set `RENDER_HTML = false` in the skill's constants block or pass `— render html: false`. To globally disable: skip with environment variable `ARIS_RENDER_HTML=0` (planned, not yet implemented; flag per skill in the meantime).\n\n## Customizing the templates\n\nThe two templates live at `skills\u002Frender-html\u002Fscripts\u002Ftemplates\u002F{academic,dashboard}.html`. Each is a single self-contained HTML file with inline CSS using `{{PLACEHOLDER}}` substitution. To customize:\n\n1. Copy one of the templates to a new name, e.g., `my_brand.html`.\n2. Edit the CSS variables in `:root { ... }` to change colors, the font stack, or layout dimensions.\n3. Add the template name to the `--template` choices in `render_html.py` `argparse`.\n4. Re-run `\u002Frender-html \u003Cinput> --template my_brand`.\n\nThe default templates are derived from the user's own academic-newspaper tutorial style (Source Serif Pro + Songti SC, 3-color palette, sticky TOC, low-flash). Stay close to that idiom for ARIS artifacts unless you have a specific reason to break the visual language.\n\n## External alternatives (for richer surfaces)\n\nFor deck \u002F poster \u002F Xiaohongshu card \u002F tweet card \u002F data report style outputs, point users to **[html-anything](https:\u002F\u002Fgithub.com\u002Fnexu-io\u002Fhtml-anything)** (Apache-2.0, 3000⭐ at time of writing). It ships 75 SKILL.md templates across 9 surfaces and detects 8 coding-agent CLIs (including Claude Code, Codex, Copilot). ARIS does *not* depend on it — `\u002Frender-html` covers ARIS-native artifacts; html-anything is the recommended path for richer publishing surfaces.\n\n## Key rules\n\n- **Do not auto-render every Markdown file.** Only artifacts on the whitelist above. File proliferation is the main anti-pattern.\n- **Do not hand-edit the generated HTML.** Edit the source, then re-render. The embedded SHA256 in the HTML meta tells you if the source has changed since render.\n- **academic-template HTML is a reviewed artifact by default**, not raw output. Cross-model Codex review (fresh thread) gates the academic deliverables — the same way `\u002Fproof-checker`, `\u002Fpaper-claim-audit`, `\u002Fcitation-audit`, `\u002Fkill-argument` gate their respective products. `--no-review` is appropriate for **interim auto-emits** (e.g., `idea-stage\u002FIDEA_REPORT.html`, per-round `review-stage\u002FAUTO_REVIEW.html`) where the source MD has already passed an upstream cross-model gate — the HTML render is then a structural conversion, not a new claim audit. For **shipped \u002F reviewer-facing \u002F audit-class** outputs, keep the full gate.\n- **The reviewer audits rendering, not research.** Claim truthfulness is owned upstream by `\u002Fpaper-claim-audit`, `\u002Fresult-to-claim`, `\u002Fresearch-review`. The HTML reviewer asks: \"did the renderer faithfully + safely convert this source?\" — nothing more.\n- **CDN dependency is opt-out, not opt-in.** Most users have internet; `--offline` is for air-gapped runs \u002F archival.\n- **The default style is academic-newspaper, not marketing-flashy.** Match the existing ARIS tonal voice. If you want decks\u002Fposters\u002Fsocial cards, point users to html-anything.\n- **Pure stdlib only.** Adding a `pip install` dependency to `render_html.py` requires an explicit decision — the helper currently has none. MCP calls live in the skill orchestrator, never in the helper script.\n",{"name":5,"description":7,"allowed-tools":24,"argument-hint":25},"Bash(*), Read, Write, mcp__codex__codex","\u003Cinput.md> [--template academic|dashboard] [--out \u003Cpath>] [--title ...] [--state \u003Cstate.json>] [--json \u003Csidecar.json>] [--offline] [--review|--no-review]","main","18193d71ea3441d7c033dbf195b8a7c75ac96f75","packages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Frender-html",0,"https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science","https:\u002F\u002Fgithub.com\u002FCZX-LBJ","一个没事喜欢随便写点代码的小菜鸡\r\n","2026-10-22 09:48:04",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add CZX-LBJ\u002Fdeepseek-science --skill render-html",{"tool":40,"command":41},"bunx","bunx skills add CZX-LBJ\u002Fdeepseek-science --skill render-html",{"tool":43,"command":44},"pnpm","pnpm dlx skills add CZX-LBJ\u002Fdeepseek-science --skill render-html","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Frender-html\u002FSKILL.md",true,[48,54,58],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"scripts\u002Frender_html.py","SCRIPT","脚本",40953,"https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Frender-html\u002Fscripts\u002Frender_html.py",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"scripts\u002Ftemplates\u002Facademic.html",21148,"https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Frender-html\u002Fscripts\u002Ftemplates\u002Facademic.html",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Ftemplates\u002Fdashboard.html",8371,"https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Frender-html\u002Fscripts\u002Ftemplates\u002Fdashboard.html",[63,70,77,84,91,98],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7FEEKSNMRMKW6JEK93PFDFY","bio-free-energy-calculations","Performs alchemical free-energy calculations including relative binding free energy (RBFE \u002F FEP+) and absolute binding free energy (ABFE) via OpenFE, FEP+, GROMACS, AMBER pmemd, and OpenMM with explicit lambda scheduling, soft-core potentials, MBAR\u002FBAR analysis, cycle-closure validation, and protocol-appropriate enhanced sampling. Compares ML alternatives (Boltz-2 affinity, DeepDock). Use when ranking analogs by binding affinity beyond docking accuracy, performing prospective lead optimization, or validating SAR predictions.","https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Fbioskills\u002Fchemoinformatics\u002Ffree-energy-calculations\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7FEE5CTE5QK55CEMXNNVWY9","bio-conformer-generation","Generates 3D conformer ensembles using RDKit ETKDGv3 with knowledge-enhanced distance geometry, MMFF94\u002FUFF force-field optimization, CREST + GFN2-xTB semi-empirical refinement, and macrocycle-aware torsion preferences. Provides explicit decision rules for single vs ensemble conformer use, RMSD pruning, energy windows, conformer count, and force-field choice. Use when preparing 3D ligands for docking, generating descriptor input for 3D QSAR, or sampling macrocycle\u002Fpeptide conformational ensembles.","https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Fbioskills\u002Fchemoinformatics\u002Fconformer-generation\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7FEDNGFEMFWD1MSEDR5H7CY","bio-admet-prediction","Predicts ADMET properties using ADMETlab 3.0 (119 platform features, including 77 prediction models with modeled-endpoint uncertainty), ADMET-AI, DeepChem MolNet, and chemprop D-MPNN with explicit handling of OECD QSAR principles, applicability domain assessment, calibration, hERG\u002FCYP\u002FAMES endpoints, and PAINS \u002F Lipinski \u002F Ro5 \u002F Veber \u002F BBB druglikeness filters. Use when filtering compounds for drug-likeness, prioritizing leads by predicted safety, or building an in-house ADMET QSAR model.","https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Fbioskills\u002Fchemoinformatics\u002Fadmet-prediction\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7FED69GFB54FRF7AV5066DN","writing-systems-papers","Paragraph-level structural blueprint for 10-12 page systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys. Provides page allocation, paragraph templates, and writing patterns. Use when user says \"写系统论文\", \"systems paper structure\", \"OSDI paper\", \"SOSP paper\", or wants fine-grained structural guidance for a systems conference submission.","https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Fwriting-systems-papers\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":19},"01M7FECVVDWZ0X0VEDTCE4B4DZ","wiki-enrich","Fill in the per-paper TODO sections of research-wiki\u002Fpapers\u002F\u003Cslug>.md pages that literature-ingest skills leave as bare scaffolds. Use when user says 'enrich wiki', 'fill paper TODOs', 'wiki body 補完', '把 paper 摘要寫進 wiki', 'research-wiki 自動填', or after a batch ingest that left papers\u002F as TODO scaffolds.","https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Fwiki-enrich\u002FSKILL.md",[],[],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":19},"01M7FECD4BCJ2MTSBDDSE3ZAJ5","web-debug-search","Search GitHub, Stack Exchange, Chinese technical communities, official documentation, and general developer web sources for software errors, compatibility problems, API usage questions, and real-world workarounds. Use for debugging and discovery only; results are not paper-citation evidence.","https:\u002F\u002Fgithub.com\u002FCZX-LBJ\u002Fdeepseek-science\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-science\u002Fassets\u002Fsources\u002Faris\u002Fskills\u002Fweb-debug-search\u002FSKILL.md",[],[]]