characteristic-voice wraps text-to-speech backends (noiz, kokoro) to inject emotional prosody into agent output. covers credential setup and runtime dependencies but omits actual synthesis steps.

talking-head-production creates synthetic videos with AI avatars and lipsync using the belt CLI. covers p-video-avatar, omnihuman, and related vendors but documentation is severely incomplete.

senpi-entrypoint provides boilerplate for onboarding an ai agent into the senpi trading platform via modular skills, with environment variable configuration and references to external error-handling docs.

autonomous-trading delegates market entry and position management to an agent given budget, target, and deadline constraints. relies on three prerequisite skills for scanning, signal detection, and stop-loss execution.

framer-motion covers disney's 12 animation principles via react declarative api, with code examples for squash/stretch and anticipation but incomplete principle coverage and vague implementation guidance.

antd covers offline querying of ant design component metadata and migration guidance via a local cli tool, bundling v4/v5/v6 reference data without network dependency.

configurator guides users through cca smart contract parameter setup via interactive prompts, with fallback to natural language if the underlying runtime lacks askuserquestion support.

affiliate-page-generator provides a framework for structuring affiliate program pages with emphasis on audience segmentation and commission transparency, but lacks concrete execution steps and assumes completion without validation checkpoints.

seo-cannibalization-detector identifies keyword overlap across pages and suggests differentiation strategies, but lacks concrete procedural steps, failure handling, and specific operational inputs/outputs.

ios-hig-design provides conceptual grounding in apple's three-pillar design philosophy (clarity, deference, depth) and names layout considerations like safe areas and dynamic island, but lacks actionable step-by-step procedures for applying these principles to real code or designs.

building-components outlines when to invoke component design guidance across primitives, accessibility, composition patterns, and publishing workflows, but lacks procedural depth and concrete failure-mode handling.

loki-mode claims multi-agent orchestration across 100+ agents but reads as aspirational rather than implemented. trigger phrase vague, procedure lacks actionable steps, edge cases absent, heavy on jargon substituting for design.

efecto-graphic-design covers asset creation via the efecto tool across multiple formats (presentations, posters, social cards, etc). lacks detailed procedures, concrete examples, and failure mode handling.

flutter-expert provides broad topical coverage of flutter 3.x architecture, state management, and multi-platform deployment, but lacks actionable step-by-step procedures, concrete trigger phrases, and failure mode guidance.

lp-agent automates liquidity provision on concentrated liquidity dexs via hummingbot api, offering command-driven setup and strategy execution without documented procedures or failure handling.

reference-signal-forms documents angular's experimental signal-based forms api with proxy-driven field traversal and schema-separated validation, but omits implementation procedures and failure modes.

scientific-slides offers broad guidance on presentation design philosophy and audience contexts but lacks concrete, step-by-step procedures for actually building slides in tools like powerpoint or google slides.

dimensional-analysis provides a workflow controller for unit and dimension tracking across numeric codebases, delegating annotation and validation to subagents. intended for defi, financial, and scientific code audits where unit mismatches cause bugs.

cro provides a conversational framework for gathering page context and performing high-level cro audits, but lacks concrete methodologies, measurement criteria, or step-by-step optimization procedures.

cloudformation-to-pulumi addresses migration from cloudformation templates to pulumi programs with emphasis on resource coverage and logical id preservation, but lacks concrete procedural steps and actionable workflows.

powerpoint covers pptx generation via html2pptx with tooling for slide auditing, reordering, and text injection. limited guidance on when to use each method or error handling.

clinical-reports covers documentation of medical information across case reports, diagnostic reports, clinical trials, and patient records. lacks procedural depth and fails to specify actionable steps or recovery strategies for common failures.

og-image-design provides a cli wrapper for converting html to social sharing images via inference.sh, with minimal guidance on og meta tags, platform specs, or branding strategy.

video-summarizer downloads videos from 1800+ platforms and generates output packages including video, audio, subtitles, and markdown summaries, but lacks actionable procedural detail and coherent error handling strategy.

gws-drive-upload handles file uploads to google drive with automatic mime type detection and optional metadata override via flags, but lacks procedural depth and failure handling.

tavily-map uses the tavily cli to enumerate urls on a website without parsing content, useful for rapid site topology discovery before selective crawling.

open-notebook is a self-hosted research tool that organizes documents and enables ai-powered analysis, podcast generation, and conversational queries with full data privacy and multi-provider llm support.

home-assistant-integration-knowledge provides structural guidelines for building ha integrations, but reads as a style checklist rather than a complete skill for implementation or review.

problem-statement provides a narrative framework for user-centered problem framing using persona, desired outcomes, barriers, root causes, and emotional impact. intended for discovery and stakeholder alignment before solution design.

phoenix-cli provides command-line access to llm observability and trace inspection, but the skill definition is incomplete, offering syntax examples without actionable procedures or error handling context.

image-to-video provides a minimal cli-based workflow for converting still images to video via inference.sh, but lacks model selection guidance, motion prompting strategy, and deployment context.

product-photography wraps the inference.sh CLI to generate product images via a bytedance diffusion model. covers studio packshots, lighting, and e-commerce conventions, but lacks procedural depth and failure handling.

cloudflare-agents enables building stateful ai agents on workers using websockets, persistent sqlite storage, and durable objects. covers multi-agent coordination, scheduling, and browser automation with rag support.

technical-roadmap-planning provides a template and checklist framework for multi-year technology strategy. lacks concrete procedures, decision criteria, and failure mode handling needed for actionable execution.

tour provides a warm welcome explaining claude's capabilities across code, memory, research, and specialized agents. it activates on capability questions and offers a structured overview with actionable next steps.

information-architecture outlines user research and structural planning for navigational design, covering card sorting, taxonomy, and validation through testing. scope is broad but lacks implementation depth.

linkedin-sales-navigator-alt describes prospect discovery from public sources without defining concrete steps, inputs, decision logic, or measurable outcomes. lacks actionable procedure.

social-media-carousel provides a single html-to-image conversion example for carousel slides. no layout rules, swipe psychology, or multi-slide orchestration implemented despite the description claiming coverage.

super-search queries past coding context from a memory store via command line, supporting personal and project-scoped lookups with basic filtering options.

newsletter-curation provides a starting point for using the belt CLI to run content search via tavily, but lacks executable guidance for actual curation, formatting, or subscriber workflows.

github-trending aggregates real-time repository and developer trending data across languages and timeframes, with analysis of project momentum and community health. lacks executable procedures and concrete trigger definitions.

dotnet-domain-modeling covers ddd tactical patterns in c# including aggregates, entities, value objects, and domain events within the domain layer, explicitly excluding persistence mapping and efcore configuration.

twitter-x-posts generates x-platform copy within character limits and thread formatting rules. covers standard and premium tiers with emoji and url accounting, but lacks actionable procedures and failure handling.

vitess covers mysql-compatible database scaling through horizontal sharding, connection pooling, and query optimization. incomplete procedural guidance limits practical application.

onboarding-cro frames user activation as a discovery process by asking contextual questions about product type and aha moments, but lacks concrete procedures for implementation or measurement.

assets exposes binance account and wallet endpoints (deposits, withdrawals, balances, trading fees, dust conversion) with comprehensive endpoint reference and credential handling guidance, but lacks concrete invocation steps and intent scenarios.

aceternity-ui catalogs 100+ animated react components for next.js with tailwind and framer motion, installable via shadcn cli. lacks actionable procedures for integration, customization workflows, or real-world usage patterns.