vc:tech-graph
Skill0 starsChanged 3 months agoMITView on GitHub
Use when you need publish-grade SVG or PNG technical diagrams for architecture, flow, sequence, UML, state, or comparison visuals, with preview used afterward for review rather than generation.
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 5 are distinctive, and 8 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Do use it for architecture diagrams, data flow diagrams, flowcharts, sequence diagrams, state machines, ER diagrams, use-case diagrams, …
- Very specific: line 19
Do use it when the output should live under process/general-plans/references/, process/general-plans/reports/visuals/, …
- Very specific: line 20
Do use preview after generation when you want explanation, HTML framing, Mermaid review, or visual self-check of the finished artifact.
- Distinctive: line 21
Do not use it as a plan or execution workflow.
- Distinctive: line 22
Do not use it to replace inline doc diagrams when a lightweight Mermaid or ASCII explanation is enough.
- Very specific: line 46
Use tech-graph when the output should become a durable artifact:
- Very specific: line 53
Use preview instead when the goal is:
- Distinctive: line 58
create inline Mermaid-first discussion material
- Very specific: line 62
Prefer .svg as the source artifact.
- Very specific: line 68
If the user does not specify an output path, route outputs to the matching the kit process folder rather than ad hoc locations like …
- Very specific: line 89
When the diagram type is ambiguous, classify it before drawing.
- Very specific: line 96
Use a template from templates/ or fixture structure from fixtures/ when it speeds up clean SVG generation.
- Distinctive: line 115
If useful, review the output visually and iterate.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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.
Files
Tech Graph
Generate publish-grade technical diagrams as SVG first, then PNG when raster export is needed.
This is a helper skill only.
- Do use it for architecture diagrams, data flow diagrams, flowcharts, sequence diagrams, state machines, ER diagrams, use-case diagrams, timelines, and comparison matrices that should become durable artifacts.
- Do use it when the output should live under
process/general-plans/references/,process/general-plans/reports/visuals/,process/features/{feature}/references/, orprocess/features/{feature}/reports/visuals/. - Do use
previewafter generation when you want explanation, HTML framing, Mermaid review, or visual self-check of the finished artifact. - Do not use it as a plan or execution workflow.
- Do not use it to replace inline doc diagrams when a lightweight Mermaid or ASCII explanation is enough.
Capability Surface
The local the kit skill now includes upstream-style assets:
- 7 visual style references
- style-to-diagram guidance matrix
- icon and SVG layout reference material
- reusable SVG templates
- JSON regression fixtures
- template-driven generation script
- batch style validation script
Load the smallest references that fit the request:
references/style-diagram-matrix.mdto choose a style and diagram familyreferences/style-1-flat-icon.mdthroughreferences/style-7-openai.mdfor exact visual languagereferences/icons.mdfor semantic shapes and product-icon guidancereferences/svg-layout-best-practices.mdfor spacing, routing, label, and export disciplinereferences/tech-graph-workflow.mdfor the kit-specific routing and proof rules
Routing Matrix
Use tech-graph when the output should become a durable artifact:
process/general-plans/references/process/general-plans/reports/visuals/process/features/{feature}/references/process/features/{feature}/reports/visuals/
Use preview instead when the goal is:
- explain a concept visually
- review an existing diagram
- generate HTML slides or recaps
- create inline Mermaid-first discussion material
Output Policy
Prefer .svg as the source artifact. Generate .png alongside it when:
- the user wants an image for sharing
- you want a visual proof artifact
- you need to inspect the render rather than only the XML
If the user does not specify an output path, route outputs to the matching the kit process folder rather than ad hoc locations like ~/Desktop or /tmp.
Diagram Types
The imported templates and references support these common families directly:
- architecture
- data-flow
- flowchart
- sequence
- comparison
- timeline
- mind-map
- agent
- memory
- use-case
- class
- state-machine
- er-diagram
- network-topology
When the diagram type is ambiguous, classify it before drawing. If the request is really an explanation-first visual, switch to preview instead of forcing a publish-grade diagram.
Workflow
- Decide if the task needs publish-grade SVG/PNG or just
preview. - Classify the diagram type and pick the output destination.
- Load the matching style reference. Default to style 1 unless the request implies another visual language.
- Use a template from
templates/or fixture structure fromfixtures/when it speeds up clean SVG generation. - Generate the SVG directly or render it with:
python3 .claude/skills/tech-graph/scripts/generate-from-template.py <template-type> <output.svg> '<json-data>'- Validate the result:
bash .claude/skills/tech-graph/scripts/validate-svg.sh <file.svg>- Export PNG when needed:
bash .claude/skills/tech-graph/scripts/generate-diagram.sh -o <file.svg>- If useful, review the output visually and iterate.
Helper Scripts
scripts/generate-diagram.sh- validates an SVG
- exports a sibling PNG when
rsvg-convertis available - warns and succeeds in SVG-only mode when
rsvg-convertis missing
scripts/generate-from-template.py- renders starter diagrams from the imported templates plus JSON structure
- supports the upstream visual style system and semantic shapes
scripts/test-all-styles.sh- renders the regression fixtures
- validates every generated SVG
- exports PNG proofs when
rsvg-convertis present - writes batch outputs to a the kit process artifact folder by default
scripts/validate-svg.sh- checks XML structure and closing tags
- uses
rsvg-convertas an extra validator when available
Dependency Contract
- Required for SVG validation: shell, Python 3, and basic Unix tools
- Optional but recommended for PNG export:
rsvg-convert
the kit allows degraded SVG-only mode when rsvg-convert is unavailable. The skill should warn clearly instead of failing silently.
Good Trigger Phrases
generate architecture diagramdraw the system flowmake this sequence diagram publishablecreate a clean SVG for this processturn this architecture into PNGgenerate a UML state machinebuild a comparison matrix for these systems
Load references/tech-graph-workflow.md for the the kit-specific routing, proof, and preview-boundary rules.
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- opencue/cuecards★ 694% same
Wrote their own version · 0
None found.
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 repository51 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- vc:kitSkill
- vc:auditSkill
- vc:autoresearchSkill
- vc:chrome-devtoolsSkill
- vc:context-engineeringSkill
- vc:debugSkill
- vc:docs-seekerSkill
- vc:generate-contextSkill
- vc:generate-planSkill
- vc:mcp-managementSkill
- vc:merge-worktreeSkill
- vc:previewSkill
- vc:problem-solvingSkill
- vc:publishSkill
- vc:repomixSkill
- vc:scoutSkill
- vc:securitySkill
- vc:sequential-thinkingSkill
- vc:setupSkill
- vc:stress-testSkill
- vc:tech-graphSkill
- vc:updateSkill
- vc:watzupSkill
- vc:web-testingSkill
- vc:xiaSkill
- vc:kitSkill
- vc:auditSkill
- vc:autoresearchSkill
- vc:chrome-devtoolsSkill
- vc:context-engineeringSkill
- vc:debugSkill
- vc:docs-seekerSkill
- vc:generate-contextSkill
- vc:generate-planSkill
- vc:mcp-managementSkill
- vc:merge-worktreeSkill
- vc:previewSkill
- vc:problem-solvingSkill
- vc:publishSkill
- vc:repomixSkill
- vc:scoutSkill
- vc:securitySkill
- vc:sequential-thinkingSkill
- vc:setupSkill
- vc:stress-testSkill
- vc:updateSkill
- vc:watzupSkill
- vc:web-testingSkill
- vc:xiaSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.

