Install
$ agentstack add skill-scibly-dev-skills-storyboard ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Storyboard
A course outline tells you what to teach. A storyboard tells you how to show it — and, just as importantly, where showing anything beyond text would be wasted effort.
This skill sits between planning (needs-analysis, didactic-reduction, scenario-generation) and production (an infographic, animation, or video-production skill). Its job is to turn a scene list into a prioritized, justified media plan that production skills can pick up directly.
Step 1 — Gather inputs
You need:
- The course/lesson structure — a scene plan, module plan, or scene list. If the user hasn't provided one, ask for it or look for
*scene-plan*.md,*storyboard*.md, or similar files in the project folder. - Production capacity — which production tools are realistically available (an infographic skill, an animation skill, a video tool like Remotion, screen recording, stock video)? If unclear, assume infographic and animation (HTML/CSS or SVG-based) production are available, since those need no extra setup.
- Constraints (ask only if not obvious) — total course duration, deadline, whether this is a pilot/MVP or a polished release.
Don't over-interview. If the scene list already has format hints (e.g. "Format: side-by-side comparison"), treat those as a starting point, not gospel — your job is to sanity-check and fill gaps, not redo everything from scratch.
Step 2 — Classify each scene by pedagogical purpose
For every scene, ask: what is this scene doing for the learner? The purpose — not the topic — determines the medium.
| Purpose | How to recognize it | Medium that fits best | |---|---|---| | Hook / emotional entry | First scene of a module; "why should I care" | Animation (narrative, before/after) | | Concept / mental model | "What is X", a metaphor, a taxonomy, "how the pieces relate" | Infographic | | Process / sequence | "How to do X step by step"; order and timing matter | Animation (progressive reveal) | | Demonstration | A tool, an interface, a real workflow in motion | Screen-recorded video | | Decision / scenario | "What would you do if...", judgment calls, risk, ethics | Interactive scenario / branching / multiple choice | | Practice / retrieval | Checking or applying what was just learned | Multiple choice, short exercise | | Summary / synthesis | End of module, "what to remember" | Infographic (recap) or plain text |
Most scenes are plain rich text/HTML — that's correct and good. A storyboard that recommends visuals for everything is a bad storyboard. The interesting work is identifying the handful of scenes where a visual genuinely changes comprehension or motivation, and being honest about the rest.
Step 3 — Decide medium vs. plain text
For each scene that isn't obviously a hook, concept, or process, run it through this check before reaching for production:
Lean toward a visual (infographic/animation/video) when:
- This is the learner's first encounter with a concept that has 3+ interrelated parts
- The content describes change over time, a sequence, or cause-and-effect
- The scene needs to create an emotional reaction (frustration, relief, urgency) to land
- Learners will want to refer back to this as a reference later (favors infographic)
Lean toward plain text/rich HTML when:
- The scene explains steps the learner will follow while doing the task (they need to read at their own pace, not watch)
- It's a transition, instruction, or orientation scene
- The content is already well-served by a list, table, or code block
- Production effort would be disproportionate to the learning gain
Within visuals, choose:
- Infographic for structure, comparison, relationships, "the big picture" — things that hold still
- Animation for sequence, transformation, before/after, "things that change" — reference your project's animation style guide, if you have one, for narrative vs. layer-reveal style
- Video only when a real interface, person, or physical demonstration is the point — animations are usually cheaper and more controllable for everything else
If the course has a guiding metaphor (in German L&D this concept is called the Leitmetapher), check whether the proposed visual reinforces it. A visual that contradicts or ignores the course's central metaphor is worse than no visual.
Step 4 — Produce the storyboard
Output two parts: an overview table, then detailed briefs for the highest-priority items.
Part A — Overview table
| # | Scene | Medium | Rationale (1 sentence) | Production | Priority |
|---|-------|--------|-------------------------|-------------|----------|
| 1.1 | The Monday-Morning Problem | Animation | Emotional hook needs movement & contrast | animation | 🔴 Must |
| 1.2 | The Three-Box Model | Infographic | Structural knowledge, 3 elements + relationships | infographic | 🔴 Must |
| 1.3 | Skill vs. Knowledge | Text + comparison table | A comparison table is enough, no added value from an image | — | — |
Priority scale:
- 🔴 Must — a core learning moment; without this medium the learning goal is at risk
- 🟡 Should — meaningfully improves comprehension or engagement, but text would still work
- 🟢 Could — nice polish if time/budget allows
- (blank/—) — plain text is the right call; don't manufacture a visual for it
Be honest about the ratio. A 15-scene course with 2–3 🔴 items and most scenes marked "—" is healthy. If everything comes out 🔴, re-apply Step 3 more critically.
Part B — Production briefs
For every 🔴 (and 🟡 if capacity allows), write a brief in this format:
### [Scene ID] — [Title]
**Medium:** Animation | Infographic | Video | Interactive
**Production tool:** animation skill | infographic skill | video tool (e.g. Remotion)
**Estimated length:** ~Xs (animation/video) or — (infographic)
**What's shown:**
[Concrete description of what appears, in what order — enough that an
animation or infographic skill could be invoked with this brief directly as input]
**Key message:**
[The single takeaway — what should stick after this plays]
**Didactic rationale:**
[Why this medium and not text — tie back to Step 3's reasoning]
Write these so they can be pasted straight into your animation or infographic production skill as the input brief.
Step 5 — Recommend production order
Close with a short, ordered punch list:
- Start with 🔴 items that establish the course's guiding metaphor or recurring visual language — everything else should look consistent with these
- Then 🔴 hooks (first scene of each module) — they have outsized impact on completion
- Then remaining 🔴 items
- 🟡 items only if the above is done and there's remaining time/budget
If the project has an existing visual style (check for a design-system file, prior infographics/animations in the output folder), note that new productions should match it — don't restate the whole design system, just point to it.
A note on restraint
The temptation with this skill is to find a reason to visualize everything — visuals are fun to plan and fun to build. Resist it. The scenes marked "—" are not failures of imagination; they're scenes where the learner is better served by reading at their own pace. A storyboard's value comes from prioritization, not from maximizing the visual count. If you're unsure whether something deserves a 🔴 or a 🟡, it's probably a 🟡 — or a "—".
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: scibly-dev
- Source: scibly-dev/skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.