agentleFS
WilliamEbongWilliamEbong/CourseForge/.claude/skills/instructional-graphics/SKILL.md

instructional-graphics

Skill0 starsChanged 56 days agoMITView on GitHub

Specifying instructional visuals as CourseForge VisualSpecs (when each of the 22 archetypes fits, which content fields to fill, label discipline, and writing genuine text equivalents). Used by storyboard authoring, visual direction, visual repair and visual reviewers.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use a figure when the idea is structural: order, cycle, branching, hierarchy, layers, comparison across dimensions, relationships, …

    line 11
  2. Very specific:

    Every VisualSpec states its purpose as the instructional job ("shows that controls act at different points between the hazard and the …

    line 13
  3. Distinctive:

    Use the course's defined terms exactly.

    line 49
  4. Distinctive:

    Keep counts legible: about 7 steps, 4 columns, 6 rows, 12 nodes.

    line 50

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Instructional graphics

When a visual earns its place

Use a figure when the idea is structural: order, cycle, branching, hierarchy, layers, comparison across dimensions, relationships, magnitude. Use prose for argument, nuance and exceptions. Never add a figure for decoration or to restate a three-item list as boxes.

Every VisualSpec states its purpose as the instructional job ("shows that controls act at different points between the hazard and the worker"), and its content does that job.

Archetypes and content fields

Content fields: items (id, label, detail, group, value), links (from, to, label), groups (id, label), columns and rows (label, cells), axes (x, y), chartType. Fill what the archetype uses; leave the rest empty or null.

ArchetypeFits whenFill
PROCESSlinear steps, start to finishitems in order; links optional (order implied)
LIFECYCLEstages that repeatitems in order; a link from last to first
FEEDBACK_LOOPan output feeds back to change an inputitems; links forming the loop, labelled with the effect
TIMELINEevents placed in timeitems with labels and dates in detail; value may hold a year
SEQUENCEhand-offs or messages between parties over timegroups = parties; items = messages with group = sender; links sender→receiver in order
STATE_DIAGRAMstates and transitions triggered by eventsitems = states; links = transitions labelled by event
DECISION_TREEyes/no or multi-way questions leading to outcomesitems = questions and outcomes; links labelled with the answer
SCENARIO_MAPa case's decision points and consequencesitems = situations, choices and outcomes; links labelled by choice
HIERARCHYparent-child levels, priority orderitems; links parent→child
LAYERED_SYSTEMstacked layers, each depending on or protecting the nextitems in order from outermost to innermost (or top to bottom) with detail
SYSTEM_ARCHITECTUREcomponents and their connectionsitems = components, groups = subsystems, links = connections
RELATIONSHIP_NETWORKmany-to-many relationshipsitems = nodes, links = relationships (labelled)
RESPONSIBILITY_MAPwho owns which dutiesgroups = parties; items = duties with group = owner; or columns = parties, rows = duties
COMPARISONtwo or three options on shared attributescolumns = options; rows = attributes with cells per option
MATRIXtwo dimensions crossedcolumns = one dimension; rows = the other; cells = the intersection
BEFORE_AFTERone situation in two statesgroups before and after; items assigned to each
CAUSE_EFFECTcauses and contributing factors leading to an effectitems = causes (group = category) and the effect; links cause→effect
CONTINUUMpositions along one scaleitems ordered from one pole to the other; axes.x names the scale
FUNNELprogressive narrowingitems in order from widest to narrowest; value optional
EVIDENCE_MAPclaims or questions and the strength of evidence for eachitems = claims with group = evidence level; groups = levels
LABELED_OBJECTparts of one objectitems = parts with detail; the object named in title
QUANTITATIVE_CHARTsourced numbers to compare or trenditems with numeric value (group for series); axes with units; chartType

Tie-breakers: a process that loops is a lifecycle; any branching is a decision tree; more than three options or more than one comparison dimension is a matrix; only sourced numbers justify a chart.

Label discipline

  • Labels are learner-facing words, 2–5 words, ideally under 30 characters. Put explanation in detail.
  • Use the course's defined terms exactly; the same concept has the same label in every figure.
  • Keep counts legible: about 7 steps, 4 columns, 6 rows, 12 nodes. Split or group beyond that.
  • Qualifications that change meaning must survive. If a label cannot carry "where reasonably practicable", the detail and text equivalent must.

Text equivalents

  • short: one sentence naming what the figure shows and its type: "Cycle of five CAPA stages, from detecting a problem to checking effectiveness."
  • long: the complete meaning, so a learner who never sees the figure learns the same thing. Write it as a list or prose that includes every item, the order or relationships, groupings, and the takeaway. For matrices and comparisons, go row by row. For decision trees, state each question and where each answer leads.
  • Describe content, not appearance. No "image of", "diagram showing boxes", "Alt text:", "Create an original…", "as shown above". Code rejects obvious authoring instructions; write so it never has to.

Colour, emphasis and interaction

Meaning never depends on colour alone: groups have labels, emphasis comes from order and text. interaction: reveal only when progressive disclosure genuinely helps and the revealed text is also in the long equivalent. interactive-data only for real datasets where exploration serves the objective.

Originality and rights

Draw original figures from the content. Never reproduce protected figures, regulator pictograms, standards diagrams or logos; describe the concept in your own structure and cite the source.

Fixing figures that fail to render

Shorten labels first, then reduce or group items, then remove links that restate obvious order. Keep the purpose, sources and archetype, and update the text equivalent to match.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.