murillodutt/cellm/skills/dse-ingest
cellm/skills/dse-ingest/SKILL.md
Ingest design decisions into the CELLM DSE from conversation context. Use when the user describes project-specific design choices, color palettes, component patterns, or UI constraints that should persist as DSE decisions.
testing
Updated Apr 16, 2026
$ install --global
skillsauthnpx skillsauth add murillodutt/cellm cellm/skills/dse-ingestInstall 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 16, 2026, 5:23 AM7.0s2 files scanned
SKILL.md
- description:
- Ingest design decisions into the CELLM DSE from conversation context. Use when the user describes project-specific design choices, color palettes, component patterns, or UI constraints that should persist as DSE decisions.
- cellm_scope:
- universal
- user-invocable:
- true
- argument-hint:
- [design decisions from conversation context]
DSE Ingest
Extract design decisions from conversation and persist them in the DSE.
Workflow
- Identify — scan conversation for design decisions (colors, spacing, component choices, layout patterns, avoid rules)
- Classify — map each decision to DSE entity type (token, component, pattern, composition)
- Search existing —
dse_searchto check if entity already exists - Update or create — use
POST /api/design-system/updateto patch the active preset with new decisions[] - Verify —
dse_searchagain to confirm decisions are indexed
Decision Types
| Type | Example Decision (ATOM format) | |------|-------------------------------| | Token | "Brand-aligned primary — warm identity (currently indigo in app.config)" | | Component | "Subtle status badges — soft for scannability (currently soft variant in app.config)" | | Pattern | "Immediate field feedback — no submit-time surprises (currently inline Zod validation)" | | Composition | "Fluid responsive cards — adapt to content, not breakpoints (currently flex flex-wrap)" |
ATOM Format (mandatory)
Each decision follows the intent + snapshot pattern:
"[INTENT — why/what] (currently [IMPLEMENTATION — how])"
Rules:
- Intent survives major version changes — it describes WHAT and WHY, never just HOW
- Snapshot in parenthetical is the current implementation — replaceable
- ATOM test: remove the parenthetical — does the sentence still guide design? If not, rewrite
- L2 only: never duplicate framework documentation (e.g., "UButton accepts variant prop")
- Use Nuxt UI v4 semantic vocabulary in snapshots:
text-dimmed,bg-muted,border-default— never raw Tailwind (text-gray-400)
Examples
"Achromatic primary — data readability over brand expression (currently 'neutral' in app.config)"
"Brand glow on hover — signature interaction (currently ring-1 ring-[var(--cellm-orange)]/30)"
"All badges monospace for data alignment (currently font-mono)"
"NEVER dark: on Nuxt UI components — color-mode is automatic"
| Bad (instruction) | Good (intent + snapshot) | |---|---| | "variant='ghost' color='neutral'" | "Subtle, non-competing actions (currently ghost/neutral)" | | "Use indigo-500 as primary" | "Brand-aligned primary (currently indigo in app.config)" | | "3-column grid on desktop" | "Fluid responsive, no rigid grid (currently flex flex-wrap)" |
Array merge behavior
POST /api/design-system/update does deep-merge. For decisions[] arrays, this APPENDS new entries. Before ingesting, dse_search the entity and check existing decisions to avoid duplicates.
Related Skills
murillodutt/cellm/skills/prose
data-ai
VerifiedTrustedCommunity
Prose override — temporarily disable quantization and respond in readable prose. Use when relational density matters, for safety-critical explanations, onboarding handoffs, or when token economy is not the priority.
SKILL.mdUpdated Apr 23, 2026
murillodutt/cellm/skills/pr-guardian
development
VerifiedTrustedCommunity
Govern explicit weekly Super PRs or maintainer-requested PR merges. Evaluates a 10-criterion readiness checklist and performs governed merge only when a user-requested PR is READY. Never creates or keeps permanent PRs. Use when: 'pr-check', 'pr-merge', 'merge this PR safely', 'is PR ready', 'guard merge', or /sk-git delegates pr-merge.
SKILL.mdUpdated Apr 23, 2026
murillodutt/cellm/skills/compress-llm
data-ai
VerifiedTrustedCommunity
Operational surface for the compress-llm Layer-1 token I/O compressor. Enable, disable, switch mode, and inspect status without editing config files. Use when tuning compression pressure for the current session or project.
SKILL.mdUpdated Apr 23, 2026
murillodutt/cellm/skills/cellm-feedback
tools
VerifiedTrustedCommunity
Generate structured upstream feedback for the CELLM engineering team. Produces evidence-first Markdown at docs/evidence/<date>-cellm-feedback-*.md for bugs, anti-patterns, deprecation gaps, and harness surprises, with optional atom registration via knowledge_ops. Use when: 'feedback for CELLM', 'send to CELLM team', 'register this as atom', 'document this anti-pattern', 'report this bug upstream'. Trigger proactively on MCP schema/runtime mismatches, mechanical edit loops (>=3 sequential edits), short deprecation windows (<6 weeks), or reusable harness surprises. Do NOT trigger for routine feature work or project-local bugs.
SKILL.mdUpdated Apr 21, 2026