Explain a complex topic back to you with visuals - the right diagram for each facet (flowchart, timeline, sequence, schema/ER, state machine, chart, comparison, mind map) plus optional conceptual illustrations - rendered in a style you control, then bundled onto a single page. SELF-CONTAINED: it explains what was already discussed or provided; it does NOT research. Decomposes the topic into the facets that most need to be SEEN, picks a visual modality per facet, renders structural diagrams via themed Mermaid (HTML->PNG, $0) and conceptual facets via /create-explanatory-image, then composes /one-pager to assemble everything into one dense page + PDF. Pick the visual style with --style (brand-dark default, clean-light, blueprint, whiteboard, custom). Use when someone says "explain this to me visually", "show me how this works", "diagram what we just discussed", "make this make sense", or wants a visual brief of a topic.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
Explain a complex topic back to you with visuals - the right diagram for each facet (flowchart, timeline, sequence, schema/ER, state machine, chart, comparison, mind map) plus optional conceptual illustrations - rendered in a style you control, then bundled onto a single page. SELF-CONTAINED: it explains what was already discussed or provided; it does NOT research. Decomposes the topic into the facets that most need to be SEEN, picks a visual modality per facet, renders structural diagrams via themed Mermaid (HTML->PNG, $0) and conceptual facets via /create-explanatory-image, then composes /one-pager to assemble everything into one dense page + PDF. Pick the visual style with --style (brand-dark default, clean-light, blueprint, whiteboard, custom). Use when someone says "explain this to me visually", "show me how this works", "diagram what we just discussed", "make this make sense", or wants a visual brief of a topic.
category
visual-communication
automation
gated
argument-hint
<topic or material to explain> [--style brand-dark|clean-light|blueprint|whiteboard|custom] [--format 16:9|a4|letter] [--max-visuals N] [--no-illustrations] [--autonomous] [--output-dir PATH]
{"version":"1.1","created":"2026-06-22T00:00:00.000Z","author":"Ability.ai","changelog":["1.1: Promoted to the public trinity-skills library — brand-dark preset now cites the vendored brand_defaults.md (swappable DEFAULT palette; the workspace's own design doc wins) instead of a private design-system path, render_html_to_png.py vendored into scripts/, --autonomous headless mode added so the skill is callable as one line, declared requires:/argument-hint + cold-start behavior","1.0 - initial: decompose topic -> pick modality per facet -> [GATE plan+style] -> render themed Mermaid + /create-explanatory-image -> [GATE visuals] -> compose /one-pager into a single page+PDF. Self-contained (no research). 4 styles + custom; brand-dark default."]}
Explain Visually
ℹ️ First, set expectations: before anything else, print one short line with this skill's version and its most recent change - the top entry of metadata.changelog above - e.g. explain-visually vX.Y - recent: <summary>. Then proceed.
Purpose
Explain a topic back to the reader with the right picture for each part of it. Where /one-pager distills text into a dense page and a diagram skill builds ONE brand-locked diagram for posting, this skill is a comprehension tool: it breaks a topic into the facets that most need to be seen, chooses the best visual modality per facet (flowchart, timeline, sequence, schema/ER, state machine, chart, comparison, mind map, or a conceptual illustration), renders each one in a style you control, and bundles the set onto a single page so you grasp the whole thing at once.
Two properties define it:
Self-contained, no research. It explains what was already discussed or provided - the current conversation, a paste, or a file. It does NOT browse, query knowledge bases, or pull external facts. If the discussion didn't establish a fact a visual needs, it marks [NEEDS: ...] and asks - it never invents.
Style is a dial. The locked brand is the default (brand-dark), but you can switch the whole visual language (clean-light, blueprint, whiteboard, or a custom brief). Every diagram in a run shares the chosen style.
The deliverable is a single bundled page (+ PDF) assembled by /one-pager, plus the individual visual PNGs kept for reuse.
When to use
Right tool when:
"Explain this to me visually" / "show me how this works" / "diagram what we just discussed" / "make this make sense"
The topic has internal structure worth seeing - a process, a sequence, a system, an evolution, a set of trade-offs
You want several complementary visuals (not one), each picked for what it shows best
You want to control the look rather than accept the locked brand
Wrong tool when:
You need ONE brand-locked diagram to post -> a diagram skill
You need a multi-slide deck -> /presentation
You need a dense text brief (mostly prose/stats, few diagrams) -> /one-pager directly
A single generative infographic is enough -> /create-explanatory-image
You need it to research a topic it doesn't already have material on -> this skill refuses; gather the material first (it is self-contained by design)
What makes this distinct
Approach
Trade-off
a diagram skill
ONE human-composed diagram, brand-locked, for posting. No modality selection, no style control, single artifact.
/create-explanatory-image
ONE generative image, iterates to correctness. Great for a metaphor; not multi-modal, not assembled.
/one-pager
Dense single page from a brief - text-first. This skill is its visual front-end: it produces the pictures, then hands them to /one-pager.
/explain-visually (this skill)
Decomposes a topic into facets, picks the right visual per facet, renders in a style you control, bundles to one page. Mostly $0 (Mermaid); only conceptual-illustration facets cost ~$0.067 each. ~1-3 min.
State Dependencies
Source
Location
Purpose
Modality guide
modality_guide.md
Facet -> visual modality -> engine; per-modality Mermaid templates; node budget - read every time
Style library
style_library.md
The 4 styles + custom: Mermaid theme vars, render-wrapper HTML, illustration prompt fragment, /one-pager theme mapping - read every time
Brand defaults
brand_defaults.md (this skill)
Read only when --style brand-dark - §0 first (swappable defaults; the workspace's own design doc wins). Exact palette/fonts the brand-dark preset cites
Render script
scripts/render_html_to_png.py
Themed Mermaid HTML -> PNG via Playwright
Prerequisites & cold start
Script paths in this file are relative to this skill's directory — run them from there,
or prefix with the directory the skill was injected into.
Requirement
Needed for
Missing →
python3 + Playwright/Chromium
rendering every Mermaid visual
Report the install command (pip install playwright && playwright install chromium); do not claim visuals that were never rendered.
Network access at render time
Mermaid + Google Fonts load via CDN
See style_library.md for offline vendoring; without either, the render fails visibly rather than silently producing an unstyled diagram.
GEMINI_API_KEY (fallback GOOGLE_API_KEY)
conceptual-illustration facets only
Not fatal — proceed as if --no-illustrations was passed and say so in the delivery report. Held by the /create-explanatory-image child skill, which fails naming the key. Pure-Mermaid runs cost $0.
Headless mode (--autonomous)
This skill is callable as one line by a schedule, an orchestrator, or another agent
(the fleet's playbook-call convention). Invoked that way there is nobody to answer a gate,
so --autonomous must be passed and both gates are skipped. In that mode:
The self-contained rule hardens: with no operator to ask, a bare topic with no material
behind it is a failed run — report it and produce nothing. It is never a licence to
research or to fill the topic from general knowledge.
[NEEDS:] gaps are dropped from the visual and listed in the final report, never guessed.
The style preset must be passed explicitly or the default applies; nothing else changes.
What to explain. If omitted, it explains what was just discussed (the current conversation + any material already in context). Can also be a topic we covered, a paste, or a file path. NEVER triggers research - if there's no substance to explain, it halts and asks.
--style
brand-dark
Visual language for ALL diagrams this run. See style_library.md. custom -> you describe the look and it maps to theme vars.
--format
16:9
Final bundled-page format (passed to /one-pager): 16:9 screen, a4/letter print.
--max-visuals
5
Cap on visuals (facets). The decomposition keeps the highest-signal facets up to this cap.
--no-illustrations
(off)
Mermaid/structural diagrams only - skip all /create-explanatory-image facets. $0, faster, fully deterministic.
--autonomous
(off)
Headless mode for scheduled/unattended runs: skip both approval gates. See Headless mode above.
--output-dir
explain_{slug}/
Where visuals/, the HTML sources, and the final PDF land. Relative to the working directory unless given absolutely.
Content model (LOCKED): explain, don't research; structure, don't fabricate
The skill's job is to re-present material that already exists in the discussion/source as visuals. The structure and teaching framing are the skill's synthesis - that's expected and good.
Facts, numbers, names, and claims must come from the source/discussion. If a visual needs a specific fact the discussion never established, leave a [NEEDS: ...] marker and surface it at Gate 1. Do not guess, and do not reach for outside knowledge to fill it.
If the source is a bare topic with no substance behind it ("explain quantum gravity") and nothing was discussed, halt and ask what material to explain. This skill does not research.
Composes
/one-pager - bundles the rendered visuals into the single page + PDF (the deliverable). Always invoked.
/create-explanatory-image - renders any conceptual-illustration facet (a metaphor / rich visual that no structural diagram captures). Invoked per such facet unless --no-illustrations.
Called by their unversioned names so child fixes propagate. Structural diagrams are rendered in-skill (themed Mermaid -> render_html_to_png.py); Mermaid is a tool, not a skill, so it is used directly.
Process (transactional)
ultrathink - the load-bearing reasoning here is decomposition + modality matching (Step 2): choosing which facets of the topic to show and the right visual type for each is what makes the explanation land. Reason carefully about what the reader actually needs to see before picking diagrams, and confirm at Gate 1 before spending any render time.
Read style_library.md (the styles, render wrapper, theme mapping).
If --style brand-dark, also read brand_defaults.md (§0 override first) for the exact palette/fonts the preset cites.
Step 1.5: Resolve flags + identify the source
Resolve --style (default brand-dark), --format, --max-visuals, --no-illustrations. Derive a {slug}; set --output-dir to {output-dir}/.
Identify the source: by default it is what was discussed before (the current conversation + material already in context). If a paste/topic/file is given, use that instead.
If --style custom, ask the user for a one-paragraph style brief (background, line/text color, accent, font feel, mood) before proceeding.
If there is no real substance to explain, halt and ask for the material. Do not research.
Step 2: Decompose the topic into visual facets
Identify the 3-6 facets (capped by --max-visuals) that most need to be seen to understand the topic. For each facet, pick a modality + engine using modality_guide.md's decision table, e.g.:
X vs Y trade-offs -> comparison (Mermaid quadrant or an HTML table)
how it all hangs together / taxonomy -> mind map (Mermaid)
a metaphor no structural diagram captures -> conceptual illustration (/create-explanatory-image)
Assign each facet a priority (must-keep / trimmable / nice-to-have). Order the facets into the reading sequence that best teaches the topic (usually: orient -> mechanism -> detail -> synthesis).
Step 3: Draft each visual's content
Mermaid facets: pick the Mermaid diagram type and write the node/edge labels - telegraphic (2-5 words). Hold the node budget (<= ~10-12 nodes per diagram; if a facet needs more, split it into two or simplify). Draft from the templates in modality_guide.md.
Illustration facets: write the concept brief you'll hand to /create-explanatory-image, plus the style's image-prompt fragment from style_library.md.
Mark any missing fact as [NEEDS: ...].
Step 4: [APPROVAL GATE 1 - PLAN + STYLE] Confirm the visual plan (skipped with --autonomous)
LOCKED first gate. No rendering, no image generation, no PDF until approved. This is where modality choices, the teaching order, the style, and any [NEEDS] gaps get settled - all free to change here.
Present the plan compactly - the topic, the style, and the ordered facet list with each facet's modality + a one-line content sketch. Example shape:
EXPLAIN-VISUALLY · {slug} · style: brand-dark · format 16:9 · bundle: /one-pager
1. FLOWCHART "How a request flows through the system"
Ingest -> Plan -> Delegate -> Verify -> Respond (5 nodes)
2. SEQUENCE "Orchestrator <-> worker handoff"
User -> Orchestrator -> Worker -> Orchestrator -> User
3. COMPARISON "Shallow vs deep agents"
axes: memory, planning, delegation, autonomy
4. TIMELINE "How we got here"
2023 prompts · 2024 tools · 2025 agents · 2026 departments
5. MIND MAP "The four pillars, at a glance" [nice-to-have]
[NEEDS] facet 4 has no dates for the 2025 milestone - confirm or drop
Then use AskUserQuestion:
"Approve this visual plan for {slug}? Want to change a modality, the order, the style, which facets, or fill the [NEEDS] gaps before I render?"
Options:
Approve - render it -> Step 5
Edit the plan -> change modality / order / facets / fill [NEEDS], re-show this gate
Change style -> switch --style (re-read style_library.md for it), re-show
Cancel -> stop, spend nothing
Once approved, write the locked plan to {output-dir}/plan.md. DO NOT render until this gate returns "Approve".
Step 5: Render the visuals in the chosen style
For each Mermaid facet (in order):
Author a themed HTML file from the render wrapper in style_library.md, injecting the style's Mermaid themeVariables + page CSS + the diagram definition. Save to {output-dir}/visuals/NN_{slug}.html.
Render at device-scale 2, using the per-modality canvas defaults from modality_guide.md:
For each illustration facet (skip if --no-illustrations):
Invoke `/create-explanatory-image` with the facet's concept brief + the style's image-prompt fragment, --output-dir {output-dir}/visuals/, matching aspect.
Verify each PNG is non-blank and nothing is clipped (see Error Recovery for the blank-Mermaid fix).
Step 6: [APPROVAL GATE 2 - VISUALS] Confirm the rendered set (skipped with --autonomous)
LOCKED second gate. No bundling until the visuals are approved. A correct plan can still render with a clipped diagram, a wrong label, or an off illustration.
Read each PNG in {output-dir}/visuals/ inline so the user sees the full set. Then use AskUserQuestion:
"Here are the {N} visuals in {style}. Approve them for the bundled page, or fix a diagram / re-style / re-order first?"
Options:
Approve - bundle the page -> Step 7
Fix a visual -> edit that diagram's labels/type (or re-run the illustration), re-render, re-show
Re-style -> switch --style, re-author the wrapper(s), re-render, re-show
Cancel -> stop, don't bundle
DO NOT compose /one-pager until this gate returns "Approve".
Step 7: Bundle into a single page - compose /one-pager
Hand the approved visuals to /one-pager as a structured block list (it styles a structured list verbatim rather than re-distilling): a header (topic title + one-line framing), then each visual as an inline-media block pointing at its rendered PNG path with a telegraphic caption, in the approved order.
Invoke `/one-pager` with that structured source, plus:
--theme {dark for brand-dark|blueprint, light for clean-light|whiteboard}
--format {--format}
--output-dir {output-dir}
/one-pager owns the single-page fit-guard + PDF assembly and runs its own content/visual confirmation on the final page - tell the user that final page approval happens inside /one-pager. Note: the bundled-page chrome follows /one-pager's brand; the visuals carry the chosen --style. brand-dark is the most coherent pairing (which is why it's the default); for other styles the diagrams keep their look while the page frame stays branded.
Step 8: Report the deliverable
Report:
The single-page PDF path (the deliverable) + its page size, from /one-pager
The visuals/ folder (each diagram PNG, reusable on its own)
The output directory
Style used, and the modality of each visual (e.g. "flowchart, sequence, comparison, timeline")
Any [NEEDS] gaps filled at Gate 1
Anything /one-pager tightened to keep it on one page
Step 9: Iterate
Change request
How to handle
Fix a diagram's label/type
Edit that visual's HTML/definition, re-render, re-bundle via /one-pager
Different style
Switch --style, re-author wrappers, re-render all, re-bundle
Add / drop a facet
Re-plan that facet (Gate 1 if adding), render it, re-bundle
Wrong illustration
Re-run /create-explanatory-image for that facet, re-bundle
Make it a deck instead of a page
Re-bundle with /presentation instead of /one-pager
Step 10: Persist
Keep the full {output-dir} (plan.md, visuals/, HTML sources, PDF) as the reproducible record.
Delivery beyond the local output directory (upload, send, publish) is the calling agent's job, not this skill's.
Completion Checklist
Source identified (conversation / paste / file); if bare, material was gathered - nothing researched, nothing fabricated
Topic decomposed into <= --max-visuals facets, each with a justified modality and a teaching order
Every [NEEDS: ...] gap filled or resolved at Gate 1
Gate 1 (plan + style) returned Approve; plan.md written
Each Mermaid visual rendered in the chosen style at device-scale 2; each illustration rendered (or --no-illustrations honored)
No visual is blank or clipped (node budget held)
Gate 2 (visuals) returned Approve
/one-pager composed with the visuals as inline-media + matching --theme; single-page PDF produced
Output dir retained; Drive upload done if requested
Final report delivered: PDF path + page size, visuals folder, style, per-visual modality
Error Recovery
Failure
Recovery
Source is a bare topic with no material
Halt at Step 1.5 and ask what to explain. NEVER research or fabricate to fill it - self-contained is the contract.
Mermaid PNG renders blank/partial
The CDN/ESM render hadn't finished at screenshot. Re-run the render (CDN now warm); if it persists, vendor mermaid@11 locally and reference it via file:// (see style_library.md).
A diagram is clipped / overflows the canvas
Too many nodes. Reduce to the node budget, split the facet into two visuals, or raise that visual's canvas height; re-render. Never ship a clipped diagram.
Mermaid syntax error (empty/garbled SVG)
Validate the diagram definition against the template in modality_guide.md; fix the syntax (often a reserved char in a label - quote it), re-render.
/create-explanatory-image fails
Retry once. If it still fails, drop that facet to a structural diagram or omit it (note the omission). Never block the deliverable on one illustration.
Playwright / Chromium render fails
Confirm Playwright + Chromium installed, retry. Report the exact error; do not bundle from stale/partial PNGs.
/one-pager can't fit everything on one page
Drop the lowest-priority visual(s) and re-bundle, or switch --format a4 for more room. The single-page guarantee is /one-pager's; respect it.
Output
Return:
The single-page PDF path (the deliverable) + page size
The visuals/ folder (individual diagram PNGs)
Output directory (plan.md, visuals, HTML sources, PDF)
Style + per-visual modality
Inline preview of the visuals and/or the final page
Any [NEEDS] filled and anything /one-pager tightened
Common issues
Symptom
Diagnosis
Fix
Diagrams all look the same flavor
Defaulted every facet to flowchart
Re-read modality_guide.md; match the modality to what each facet actually shows (timeline for evolution, sequence for handoffs, etc.)
Visual reads as a "wall of nodes"
Over the node budget
Split into two diagrams or simplify; <= ~10-12 nodes each
Style looks off-brand for brand-dark
Didn't read design_system.md
brand-dark cites the locked palette/fonts - read it and use the exact values
Page chrome clashes with a non-brand style
/one-pager chrome is brand-locked
Expected; the visuals carry the style. Use brand-dark for full coherence, or accept frame/visual divergence
It tried to look something up
Ignored the self-contained rule
Stop. Explain only the discussed/provided material; mark gaps [NEEDS]
Related skills
/one-pager - the bundler this composes; use it directly when the content is text-first with few diagrams.
A diagram skill (if installed) - one brand-locked diagram for posting (not multi-modal, not style-switchable).
/create-explanatory-image - one generative infographic; composed here for conceptual facets.
/presentation - swap in for /one-pager when you want a deck instead of a single page.
