| name | paper-poster-html |
| description | "DEFAULT poster pipeline โ build an academic conference poster (ICML/NeurIPS/ICLR/CVPR/...) as a single HTML/CSS file with measurement-driven hard gates, real paper figures, a two-hue design-token system, and print-ready PDF via headless Chromium. Use when the user says \"ๅๆตทๆฅ\", \"poster\", \"conference poster\", \"paper poster\", or asks to design/redo a research poster." |
Paper Poster (HTML): measurement-gated poster generation
One HTML file styled for an exact print canvas (@page { size: W H }), rendered to PDF
via Playwright print emulation. Iterate by measuring, not eyeballing โ the screen
preview lies; only print emulation at the correct viewport tells the truth. Core gate
machinery is adapted from posterly (MIT, ยฉ
2026 Ruishuo Chen โ see NOTICE.md and LICENSES/posterly-MIT.txt in the mainline
skill directory); ARIS adds style discipline gates, figure-provenance gates, the
cross-model review loop, and the anti-patch-loop fix vocabulary.
Why this skill exists (the failure it prevents)
A predecessor pipeline produced a poster with 30+ colors, zero real paper figures, a
screen-pixel canvas, and tiny formulas floating in oversized boxes, then spent 12+
review rounds making it worse โ each round added a new badge color or bespoke SVG
patch. The cure is structural, not exhortative:
- Hard gates run before any aesthetic opinion (alignment, style, assets must PASS
first โ a reviewer never sees an unmeasured poster).
- A closed fix vocabulary โ visual-review fixes can only touch design tokens,
whole catalogued components, content rebalance, assets, or canvas choice. New inline
styles / new hex values / bespoke decorations are structurally forbidden.
- Two-hue discipline as a machine check, not a style suggestion.
- Real paper figures with provenance manifest, or the gate fails.
Mental model
paper (.tex / PDF) โโโบ content plan + claimโevidence audit (fresh reviewer agent)
โ
figures extracted โโโโโโโโโโค FIGURE_MANIFEST.json (provenance, sha256)
(real paper figures ONLY) โผ
template scaffold โโโบ fill โโโบ run_gates.py โโโโ HARD, loop here
preflight โ style โ asset โ measure โ polish
โ all hard gates PASS
โผ
executor visual review (โค3 issues ร โค3 rounds, fix-vocabulary only)
โ score โฅ 9
โผ
final cross-model review (fresh reviewer agent, full HTML+PDF)
โ pass
โผ
verify-final โ poster.pdf + GATE_REPORT.json
Constants
-
SKILL_SCRIPTS โ helpers and templates are single-owner and ship inside the
mainline skill (Arch C) at skills/paper-poster-html/scripts/ and
skills/paper-poster-html/templates/. Resolve them in this order:
SKILL_HOME=""
[ -d ".agents/skills/paper-poster-html/scripts" ] && SKILL_HOME=".agents/skills/paper-poster-html"
if [ -z "$SKILL_HOME" ] && [ -z "${ARIS_REPO:-}" ] && [ -f .aris/installed-skills-codex.txt ]; then
ARIS_REPO=$(awk -F'\t' '$1=="repo_root"{print $2; exit}' .aris/installed-skills-codex.txt 2>/dev/null) || true
fi
[ -z "$SKILL_HOME" ] && [ -d "skills/paper-poster-html/scripts" ] && SKILL_HOME="skills/paper-poster-html"
[ -z "$SKILL_HOME" ] && [ -n "${ARIS_REPO:-}" ] && [ -d "$ARIS_REPO/skills/paper-poster-html/scripts" ] && SKILL_HOME="$ARIS_REPO/skills/paper-poster-html"
[ -z "$SKILL_HOME" ] && [ -d "$HOME/.codex/skills/paper-poster-html/scripts" ] && SKILL_HOME="$HOME/.codex/skills/paper-poster-html"
if [ -z "$SKILL_HOME" ]; then
echo "ERROR: paper-poster-html scripts not resolved. Re-run the ARIS Codex install." >&2
fi
SKILL_SCRIPTS="$SKILL_HOME/scripts"
If unresolved, the install is broken: abort and tell the user to re-install (Policy A
โ the gates ARE the skill; never improvise replacements).
-
REVIEWER_MODEL = gpt-5.5, reasoning effort xhigh, fresh reviewer agent per
review call (a new spawn_agent: every time; never reuse a reviewer agent across
review boundaries).
-
CANVAS โ from the venue's official spec, looked up live in Phase 0. Never assume.
(Known anchor: ICLR 2026 main = 185ร90 cm landscape per its official printing
service; ICML/NeurIPS commonly 60ร36 in landscape; workshop posters often 61ร91 cm
portrait. Specs change yearly โ verify.)
-
PALETTE โ default = templates/tokens/generic.json (slate-blue #2D5F8B accent
- gold
#C9A24A highlight + neutrals) for all venues. Venue packs are opt-in via
โ venue-colors: true. Purple-dominant accents (hue 250โ285) are banned unless the
user passes โ allow-purple: true.
-
AUTO_PROCEED = false โ wait for explicit confirmation at every ๐ฆ checkpoint.
-
OUTPUT_DIR = poster_html/ in the working directory.
Workflow
Phase 0 โ Resume, dependencies, venue spec
- Resume: if
poster_html/POSTER_STATE.json exists with status: in_progress
(< 24 h), resume from the saved phase.
- Dependencies (degradation chain, in order):
- Playwright + bundled Chromium โ if missing,
python3 -m playwright install chromium โ if install fails but system Chrome exists, scripts fall back to
channel="chrome" โ if all fail: you may produce the content plan and scaffold
only, label everything "not print verified", and must NOT emit a final PDF.
If print rendering, PNG review, or PDF verification is impossible in the current
environment, stop and tell the user what to configure. Do not silently degrade
this skill into an unmeasured text-only poster draft.
pdfinfo missing โ PyMuPDF reads PDF dimensions. At least one of
pdftoppm / PyMuPDF must exist for PNG review renders.
- MathJax: download
tex-svg.js once into poster_html/assets/mathjax/ and
reference it locally in the HTML. CDN is acceptable only for drafts; the measure
gate hard-fails on unrendered MathJax either way.
- Venue spec lookup (live): consult the venue's official poster-instructions page
(search + fetch). Extract dimensions, orientation, font floor, logo policy,
anonymity rules, file format. Record
{spec, source_url, retrieved} into
POSTER_STATE.json โ specs change yearly; never reuse a cached spec silently.
๐ฆ Checkpoint: echo the venue spec table (canvas, orientation, source URL) and the
chosen template. Wait.
Phase 0.5 โ Design discovery (one question batch)
Ask the user once, โค4 questions: layout template (from templates/README.md), palette
(default generic pack / venue pack / custom within constraints), logos + venue mark
(paths or "none" โ never fabricate; check the venue's logo policy), QR target (paper /
code / project page / none โ generate offline with qrencode or python-qrcode;
never a remote QR-service URL). Persist answers in POSTER_STATE.json as
design_decisions โ re-read before any later "improvement" so deliberate choices are
never reverted.
Phase 1 โ Paper ingest, content plan, claim audit
-
Read the paper source (.tex ideal; PDF otherwise). Extract: title/authors/affils,
the 3โ5 headline numbers, core method (equations verbatim), main results
(tables/figures and what they show), takeaways. Build
poster_html/POSTER_CONTENT_PLAN.md โ what goes in which column, word budget per
card. Target density (excluding table cells, captions, author line, footer):
standard poster 550โ850 words; dense theory+empirical poster 750โ1050 words,
allowed only when โฅ2 compact components are used (eqn-anatomy, flow-strip,
derived-col, claim-pills, keybox--4). Warn yourself below 500 words on a
4-column landscape (it will read as sparse next to professionally dense posters)
unless the template is hero/visual-first; warn above 1100 unless the user asked for
dense mode. Bullets โค 8 words when possible โ density comes from structure, not
long prose. Prefer compact structure over prose: if the paper contains an
explicit objective, algorithm, theorem mechanism, or baseline comparison, extract at
least two of: (1) empirical objective / loss stack; (2) term-by-term equation
anatomy; (3) a method-flow strip grounded in paper variables; (4) a derived-ฮ column
for method-vs-baseline rows; (5) a 4-up implementation/theory keybox; (6) a
claim/evidence pill table for numeric-heavy posters. Do not invent an algorithm.
If the paper has only an objective, label the component "objective flow" or "loss
anatomy", never "algorithm".
-
Cross-model content audit (fresh reviewer agent): give it the content plan path
- paper source path(s) โ paths only, no summaries โ and ask for a claimโevidence
table. Save to
poster_html/CLAIM_EVIDENCE.md.
spawn_agent:
model: gpt-5.5
reasoning_effort: xhigh
message: |
Audit a conference-poster content plan against its source paper.
Read these files yourself (no other context is provided):
- poster_html/POSTER_CONTENT_PLAN.md
- [paper source path(s)]
For EVERY claim, number, equation, and attribution in the plan, output one row:
| claim on poster | paper file:line | paper says (verbatim) | match? |
with match โ {OK, NUMERIC-MISMATCH, OVERCLAIM, MISSING-PRECONDITION,
NOT-IN-PAPER, SCOPE-NARROWED}. End with a count per category.
-
Fix every non-OK row or record it as a user-acknowledged tradeoff.
๐ฆ Checkpoint: content plan + audit summary. Wait.
Phase 2 โ Real paper figures (provenance-gated)
Source preference chain:
- Paper source
figures/ (vector SVG/PDF โ convert to SVG via
inkscape/pdf2svg if available, else rasterize โฅ 2ร rendered px).
- PDF-only:
extract_pdf_figures.py contact-sheet + auto to list candidate
regions โ pick crops (๐ฆ human confirms crop choices) โ crop at 300โ450 DPI.
- Last resort: user supplies explicit
page,x0,y0,x1,y1 bboxes.
Then preprocess_figures.py --autocrop every asset. Every paper-derived image gets a
FIGURE_MANIFEST.json entry (source hash, page, bbox, dpi, sha256, natural_px) and is
embedded as <img data-source="paper" data-asset-id="...">.
Hard rule: โฅ 2 paper-derived visuals or the asset gate fails. Theory-only papers
may waive the total-area rule (--waive-total-area) at a human checkpoint โ never
silently. Never draw bespoke decorative SVG "figures" as substitutes.
Figure-area bands (asset gate, fractions of body): total target 14โ22 %
(warn < 12 % / > 24 %, hard < 10 % / > 28 %); per ordinary figure target 4โ8 % (warn
10 %, hard > 13 %); figure--duo combined 8โ12 %. Hero templates pass --hero
(centerpiece may take 30โ40 %). The failure mode is symmetric: too small reads as
decoration, too big crowds out content. Sibling figures that share axes or tell a
beforeโafter story belong in one figure--duo card, not two cards.
Phase 3 โ Scaffold + tokens
cp "$SKILL_HOME/templates/<chosen>.html" poster_html/poster.html; retarget @page
.poster dims to the venue canvas (two edits, same values); apply the chosen token
pack onto the :root DESIGN TOKENS block; fill content per the plan; embed manifest
figures. Run preflight + style_check โ both must PASS before any layout iteration.
(A fresh scaffold is expected to fail measure โ that gate judges a filled poster.)
Phase 4 โ Layout hard loop
After every layout change:
python3 "$SKILL_SCRIPTS/run_gates.py" poster_html/poster.html \
--tokens <pack.json> --manifest poster_html/FIGURE_MANIFEST.json \
--report poster_html/GATE_REPORT.json
Canonical order: preflight โ style โ asset โ measure โ polish. Targets: column-bottom
spread < 5 px (aim < 3), footer gap โ [30, 50] px, intercard gap โ [12, 50] px,
canvas-fill โ [95, 101] %, poster bbox aligned to page within ยฑ2 px. Fix guidance for
each failure mode lives in the gate output and templates/COMPONENTS.md. Do not
proceed while any hard gate fails. Do not let a reviewer see an unmeasured poster.
Balance under-filled columns with content from the paper (Gate C), never with
whitespace, space-between, or stretched cards.
Phase 5 โ Executor visual review (gated aesthetics)
Render and read the result yourself:
python3 "$SKILL_SCRIPTS/render_preview.py" poster_html/poster.html
pdftoppm -r 100 poster_html/poster_preview.pdf poster_html/review_full -png -f 1 -l 1
Score strictly 1โ10. Critical caps: < 2 real paper figures โ โค 3; broken canvas /
clipped content / unreadable math โ โค 4; โฅ 4 visible hue families or gradient-heavy
header โ โค 4; large blank cards or columns โ โค 5; fabricated visual claim โ โค 3.
Checks: posterly-showcase gestalt (would this hang next to a professionally designed
poster without looking like a patched dashboard?), single-accent discipline, real
figures readable and central, print hierarchy (title โ headline stats โ figures โ
detail), column fill, equation prominence (no tiny math in oversized boxes),
serif-body/sans-display pairing, no gradient kitsch, component consistency, 60-second
narrative. Output format:
SCORE: N/10
CAPS_TRIGGERED: ...
TOP_ISSUES: (max 3)
ALLOWED_FIX_TYPE per issue: token | component | rebalance | asset | template/canvas
PATCH_LOOP_RISK: low | medium | high
Loop: fix (fix vocabulary below) โ re-run Phase 4 gates โ re-score. โค 3 issues per
round, โค 3 rounds. Score โฅ 9 โ Phase 6. Still < 9 after 3 rounds โ STOP patching;
escalate to template / canvas / content re-choice (back to Phase 3) or a human
decision. Never enter round 4 of cosmetic patching.
Fix vocabulary (closed set โ the anti-patch-loop core)
Allowed: (a) edit a :root token value; (b) swap/remove/add a whole component
instance from templates/COMPONENTS.md; (c) content rebalance (move a card across
columns, trim/grow text from the paper, resize a figure within its AR band);
(d) template/canvas re-choice; (e) global edits to an existing component's CSS
that reference only tokens; (f) switching predefined variants (.eqn--large,
.card--compact, .figure--wide, .nowrap, โฆ); (g) asset fixes (re-crop, swap
for a clearer figure from the same paper, re-preprocess).
Forbidden: new inline styles, new hex values anywhere, bespoke decorative SVG,
per-element font-size overrides. A new component may not be born inside the visual
loop โ stop, get a human checkpoint, add it to COMPONENTS.md, re-run from Phase 3.
Phase 6 โ Final review (fresh reviewer agent, cross-model)
All hard gates PASS + polish warnings zero-or-waived + visual โฅ 9 first. Then a fresh
reviewer agent reviews the final artifacts (not the content plan) โ paths only, no
executor framing:
spawn_agent:
model: gpt-5.5
reasoning_effort: xhigh
message: |
Final print-readiness audit of a conference poster. Read these files yourself:
- poster_html/poster.html (final single-file poster)
- poster_html/poster_preview.png (rendered preview โ view it)
- [paper source path(s)]
- poster_html/GATE_REPORT.json
- poster_html/CLAIM_EVIDENCE.md
Check: (1) fidelity & overclaims RE-CHECKED on the final text (polish introduces
new claims), (2) residue (\ref{, TODO, raw < in math, missing images, remote
URLs), (3) visual rhetoric (headline numbers prominent, banner readable from
2 m), (4) gate-log coherence.
Verdict: PRINT-READY or NEEDS-FIX with a numbered, severity-ordered issue list.
The reviewer recommends; it does not edit. Any fix โ back through Phase 4/5 gates โ
never straight to re-review.
Phase 7 โ Final verification + report
python3 "$SKILL_SCRIPTS/poster_check.py" verify-final poster_html/poster_preview.pdf \
--from-html poster_html/poster.html --max-size-mb 20
Page count 1, dimensions match @page, size โค 20 MB, no TODO/residue, no remote
assets. Report: PDF path, final spread px, footer-gap range, gate summary table,
unresolved waivers, reviewer verdict. Update POSTER_STATE.json โ done.
State persistence
poster_html/POSTER_STATE.json: {phase, venue, canvas{w,h,orientation,source_url, retrieved}, template, token_pack, design_decisions{...}, figures_selected[], visual_rounds, reviewer_verdicts{audit, final}, status, timestamp} โ written after
every phase; enables compact-recovery resume.
Key rules
- Measure, don't eyeball. No layout claim without
run_gates.py output.
- Gates before aesthetics. A reviewer only ever sees a poster whose hard gates
PASS. This ordering is what kills the patch-loop death spiral.
- Never invent paper numbers or figures. Numbers come from the paper source;
visuals carry manifest provenance. Fabrication = critical cap โค 3.
- Two hues, one system. Accent + gold + neutrals. The style gate enforces it;
don't negotiate with the gate.
- Real figures are the poster. A poster without the paper's own figures is a
dashboard, not a poster.
- Fix vocabulary is closed. If a fix isn't expressible as token / component /
rebalance / asset / canvas, it's the wrong fix.
- Cross-model verdicts. The executor drives the loop and scores visuals;
acceptance of content fidelity comes from the fresh reviewer agent (a loop can
drive, never acquit).
- Preserve user decisions. Re-read
design_decisions before "improving" anything.
- Vendor boundary.
poster_check.py, render_preview.py, _posterly/ are
vendored from posterly โ keep diffs minimal; ARIS-side logic goes in the new
scripts, not in vendored files.
Review tracing
Save every reviewer call's trace per ../shared-references/review-tracing.md to
.aris/traces/paper-poster-html/<date>_run<NN>/ (audit + final agents, raw responses).
Output contract
poster_html/
โโโ poster.html # single-file source of truth
โโโ poster_preview.pdf # print-emulated, verify-final-checked
โโโ poster_preview.png # thumbnail
โโโ POSTER_STATE.json # resume state
โโโ GATE_REPORT.json # canonical gate ledger (schema v1)
โโโ POSTER_CONTENT_PLAN.md # what-goes-where + word budgets
โโโ CLAIM_EVIDENCE.md # reviewer claimโevidence audit
โโโ FIGURE_MANIFEST.json # figure provenance (sha256, page, bbox, dpi)
โโโ assets/{paper_figures,logos,qr,mathjax}/
When NOT to use
- Slides, not a poster โ
/paper-talk / /slides-polish.
- The paper's headline isn't stable yet โ fix the paper first; a poster amplifies
whatever story it's given.