Test each boundary: - Quant-dequant roundtrip: quantize the fp32 input to INT8, then immediately dequantize back to fp32.
Create helpful, actionable zero-content states
Guides the design, authoring, and debugging of predictable Claude Code skills using proven architectural principles.
Design React components from Figma files.
Review or audit a design/UI across 6 weighted dimensions with Nielsen's 10 heuristics and a prioritized findings table.
Generate, extend, or audit design tokens in DTCG format with the 3-tier architecture (primitive → semantic → component).
Comprehensive guide to designing reliable, scalable data systems covering databases, streaming, and consistency
Generates and manages custom Claude Code skills, hooks, agents, and slash commands to streamline development workflows.
Create and manage Claude Code skills following Anthropic best practices.
This skill should be used when the user asks to create a new skill, repair an existing skill, improve trigger descriptions, reorganize…
Standardizes the creation and formatting of Claude Code skills using a comprehensive markdown blueprint.
Skill Dropshipping Product Launcher
Model management for dx_stream pipelines
Classify whether user intent was satisfied in a Datadog Agent Observability trace or session.
Assess skill effectiveness from relevant agent history or scoped execution evidence.
Evolves skills based on usage patterns. Use when improving or rolling back skill definitions.
The extracted context is automatically formatted into the next prompt: Uses ContextExtractor from the meta-prompting engine:
Keep Figma and code in sync — map the 3-tier DTCG tokens to Figma Variables (collections + modes), sync in either direction, use the Figma…
Validate skill/command file format and structure.
Audit workforce skill gap identification systems and labor market alignment.
Creates, updates, and fixes Agent Skills end-to-end: drafts new skills from a short interview, improves existing ones against Anthropic…