grazianoguiducci/diagram-generator
plugins/d-nd-core/skills/diagram-generator/SKILL.md
Generate conceptual diagrams from article content. Trigger when the operator says 'genera diagramma', 'diagram', 'visual spec', 'conceptual map', or when a page is created/updated and needs a structural visualization.
$ install --global
skillsauthnpx skillsauth add grazianoguiducci/d-nd-seed diagram-generatorInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 22, 2026, 8:02 PM14.6s1 file scanned
SKILL.md
- name:
- diagram-generator
- description:
- Generate conceptual diagrams from article content. Trigger when the operator says 'genera diagramma', 'diagram', 'visual spec', 'conceptual map', or when a page is created/updated and needs a structural visualization.
- user-invocable:
- true
Diagram Generator — From Content to Structure
When an article is created or updated, this skill generates a conceptual diagram that captures the logical structure: key concepts, directional relations, and contextual copy for each node.
When to use
- A new article is published and needs a visual structure
- An existing article is updated and the diagram is stale
- The operator asks for a conceptual map of any content
- The admin panel "Generate Diagram AI" button is clicked
How it works
- Read the article content (title + body)
- Extract 3-7 key concepts as nodes
- Identify relations: cause→effect, sequence, dependency, tension
- Generate contextual copy for each node (observer positioning: when → content → result)
- Produce a DiagramSpec JSON compatible with the renderer
Two modes
LLM mode (default): sends content to LLM API. Best results — understands narrative.
Requires: GODEL_API_KEY + GODEL_API_URL (or DIAGRAM_API_KEY + DIAGRAM_API_URL)
Structural mode: rule-based extraction from headers and text patterns. No API needed. Use when: offline, quick generation, or API unavailable.
CLI
# LLM mode
python diagram_generator.py --content article.md --json
# Structural mode
python diagram_generator.py --content article.md --structural --json
# From stdin
cat article.md | python diagram_generator.py --stdin --title "Article Title"
As library
from diagram_generator import generate_diagram, generate_diagram_structural
spec = generate_diagram('Title', 'Content...', lang='it')
# spec = {type, entities: [{id, label, color, context, link}], interactions: [{from, to, type}]}
Copy rules for nodes
Every node's context field positions the observer:
- When/where this matters (situation)
- What happens (content)
- What comes out (result)
Inclusive language: "we/our" not "you/tu". The visitor is on the same boat.
Iteration
The first generation is a draft. Review, edit labels and relations, test with a fresh eye. 4 clear nodes > 7 cluttered nodes.
Complete guide
See DIAGRAM_GENERATOR_GUIDE.md in plugins/d-nd-core/scripts/.
$ARGUMENTS
Eval
Trigger Tests
"genera diagramma per questo articolo" → activates
"genera un diagramma concettuale" → activates
"visual spec per la pagina" → activates
"come funziona il deploy" → does NOT activate
Fidelity Tests
Given article with 5 sections → produces 5 entities + 4 flow interactions
Given article with cycle keywords → last interaction points to first entity
Given empty content → returns error or empty spec
Context copy always inclusive (no "tu/you")
Related Skills
grazianoguiducci/third-act
testing
VerifiedTrustedCommunity
Closure reflection protocol. After a significant work block concludes (feature shipped, session ending, major commit landed, cross-node coordination resolved), runs a 10-question interview that extracts meaning, impact, and next questions — then emits multiple audience-specific artifacts (changelog, external editorial, AI integration docs, memory crystal, backlog seed). Turns implicit maturation into explicit narrative. Use at the end of meaningful work, not after trivial edits.
1SKILL.mdUpdated May 2, 2026
grazianoguiducci/sieve-orchestrator
testing
VerifiedTrustedCommunity
The neutral form of the D-ND method. Meta-skill that recognizes context and orients toward the right specialization (cec, autologica, cascade, assertion-verifier, etc.). Activate at the start of a non-trivial work block or when input matches trigger words ('where are we', 'what here', 'orchestrate', 'connect', 'sieve this').
1SKILL.mdUpdated May 2, 2026
grazianoguiducci/publish-safe
development
VerifiedTrustedCommunity
Five mechanical gates for any content publish pipeline with CMS + rendering layers. Prevents false security: 'API returned 200' does not mean 'visitor sees clean content'. Use when writing content to a multi-layer serving system (CMS API, static files, prerendered HTML, cached copies).
1SKILL.mdUpdated May 2, 2026
grazianoguiducci/poly-consult
testing
VerifiedTrustedCommunity
Multi-node consultation protocol for high-leverage decisions. Dispatches the same question to N independent LLM/agent nodes in isolation, then synthesizes their responses into a summa that exposes convergence (high-confidence claims), dissensus (real uncertainty zones), and emergent points (insights no single node produced). Reduces single-node training bias. Supports recursive escalation for stable-state convergence. Use for decisions that propagate via A14 cascade — seed updates, crystallizations, advisory→mechanical promotions, high-visibility copy, lab result interpretation.
1SKILL.mdUpdated May 2, 2026