flowchart
Skill2 starsChanged 2 months agoMITView on GitHub
Create editable flowcharts, system maps and funnels with semantic icons, structured operation cards, distinct skill and decision shapes, and verified routing.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- Ankh2454/flowcharttheir own version, first seen Nov 1, 2026
- Whatscheap/igly-canvas-plugintheir own version, first seen Oct 17, 2026
- Unclecheng-li/AI_Animationtheir own version, first seen Aug 18, 2026
- jelbirt/LLM-Skill-Buildtheir own version, first seen Aug 6, 2026
- danshapiro/skill-flowcharttheir own version, first seen Jul 5, 2026
- shiruijun/claude-skilltheir own version, first seen Jun 11, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- Unclecheng-li/AI_Animation★ 1.7ksame name
- danshapiro/skill-flowchart★ 4same name
- Ankh2454/flowchart★ 1same name
- jelbirt/LLM-Skill-Build★ 1same name
- inancgumus/skills★ 0same name
- shiruijun/claude-skill★ 0same name
- Whatscheap/igly-canvas-plugin★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository54 other files it gives its agents.
- AGENTS.mdAGENTS.md
- context/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- context/CLAUDE.mdCLAUDE.md
- brand-reviewSkill
- brand-systemSkill
- capture-contextSkill
- content-strategySkill
- crm-presync-checkSkill
- graphics-designerSkill
- gtm-startSkill
- launch-videoSkill
- lead-magnet-creatorSkill
- lead-sourcingSkill
- linkedin-copywriterSkill
- long-formSkill
- manim-videoSkill
- newsletter-writerSkill
- outbound-copywriterSkill
- oxygen-dedupeSkill
- oxygen-deliverabilitySkill
- oxygen-diagnosticsSkill
- oxygen-email-infraSkill
- oxygen-gtmSkill
- oxygen-knowledgeSkill
- oxygen-linkedin-marketingSkill
- oxygen-quickstartSkill
- oxygen-recipesSkill
- oxygen-sequencer-enrollSkill
- oxygen-sequencerSkill
- oxygen-table-tidySkill
- oxygen-uniboxSkill
- oxygen-workflow-authoringSkill
- personalize-ai-columnSkill
- qmdSkill
- repurpose-contentSkill
- researcherSkill
- setup-workspaceSkill
- shorts-audioSkill
- shorts-cutSkill
- shorts-editSkill
- shorts-motionSkill
- shorts-qaSkill
- signals-sweepSkill
- spin-tagsSkill
- split-test-designerSkill
- video-useSkill
- voice-calibrationSkill
- week-postsSkill
- x-copywriterSkill
- youtube-descriptionSkill
- youtube-publisherSkill
- youtube-scriptSkill
- youtube-thumbnailSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 11
Read AGENTS.md, the actual workflow sources and brand/BRAND.md.
- Very specific: line 17
Use narrowing shapes for filtering, lanes for ownership and arrows for real dependencies.
- Distinctive: line 23
Read the visual grammar and helper API when authoring a scene.
- Distinctive: line 32
Use semantic line icons for recognizable concepts such as a source file, person, company, context brain, calendar, message or verification.
- Distinctive: line 34
Use approved client tokens when available.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Workspace: all context paths below (identity, audience, strategy, voice, brand, raw, output, index, log, workspace preferences and context scripts) resolve from context/ in the kit root. Read both the root AGENTS.md and context/AGENTS.md. Skill-local references/ and flowchart scripts/ resolve from this skill directory.
Draw an editable flowchart
Read AGENTS.md, the actual workflow sources and brand/BRAND.md. Resolve paths from this skill's repository. For a content funnel, read the strategy and stage pages. When editing, inspect the existing scene and generator, preserve manual changes and retain the current viewing URL.
Model the process
Identify triggers, inputs, operations, decisions, outputs, exceptions and feedback before choosing the layout. Give each nontrivial branch an explicit destination.
Use narrowing shapes for filtering, lanes for ownership and arrows for real dependencies. A taxonomy of pillars or test dimensions is a set of choices, not a sequential process. Top/middle/bottom describe content jobs; readers need not pass through every stage.
In signal-based outreach, distinguish engagement capture, buyer-fit qualification, routing, draft creation, review and sending. Label proposed steps and dated snapshots. Engagement alone does not imply buyer intent, permission or a sale. Drawing an operation does not execute it.
Make the roles visible
Read the visual grammar and helper API when authoring a scene.
- Source/reference: quiet file card or compact list; no action-style header that implies execution.
- Function/operation: square card, light title band, semantic icon, labeled input/work rows and a distinct output strip.
- Skill: rounded card and contrasting title band for a procedure that applies context or judgment. Classify its responsibility, not whether its name contains AI.
- Decision: short question inside a diamond, with labeled exits and explicit hold/retry destinations.
- Signal/result: a compact event or result shape. A numeric result can use large type with its unit, population and date.
- Grouping: shared alignments or a light container for a meaningful stage. Use different compositions for an overview, taxonomy, funnel and result breakdown.
Use semantic line icons for recognizable concepts such as a source file, person, company, context brain, calendar, message or verification. Keep icon size and stroke consistent. Each icon stays editable and grouped with its card. Icons are not a substitute for readable labels.
Use approved client tokens when available. The included helper defaults to neutral grays and a standard font slot; it contains no logo, signature, brand palette or client connections. Do not copy a reference client's visual identity or domain content.
Build and inspect
Use the named tool or existing project canvas. For Excalidraw, scripts/flowchart.py and its adjacent primitives.py build native shapes, semantic icons, grouped cards, embedded images and reciprocal arrow bindings using Python's standard library. No scene is created on import. Pass a theme and font slot explicitly when integrating with a custom canvas.
For other tools, reuse the visual grammar in their actual native model. Lucide is an icon library, not a diagram editor. Do not claim an SVG is a native hosted diagram.
Compute connector endpoints from current card bounds. Route long returns outside the main flow. Leave room for labels. Shorten, wrap or enlarge a card before shrinking its text.
Python estimates text sizes for authoring; actual glyph measurement and visual inspection are still required. Render or import using the available editor. Verify actual fonts, text containment, connector collisions, branch meaning, reciprocal bindings and image aspect ratio. Do not silently substitute fonts or flatten editable elements to make validation pass.
Open the rendered overview and useful detail views. A dense working map needs zoom; a phone preview checks hierarchy, not proof that every field is readable. Split a social graphic into fewer nodes or multiple slides.
Deliver the editable scene, preview where available, dimensions, viewing URL and any inspection limitation. Keep reusable mechanics here and domain content in the diagram's own folder.
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
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.

