openclaw-rl-training integrates asynchronous reinforcement learning with live conversation interception to optimize ai agent policies without blocking user interactions.

apple-hig-expert frames itself as a senior design lead offering guidance on apple interface standards, liquid glass aesthetics, and accessibility across iOS, macOS, watchOS, and visionOS. lacks concrete procedural steps and testable outputs.

language-evolution establishes principles for realistic linguistic development in fictional worlds, covering phonological change, social stratification, and contact effects, but lacks procedural depth and actionable workflow.

homelab-vlan-segmentation covers network isolation patterns for home environments using UniFi, pfSense/OPNsense, and MikroTik. segregates iot, guest, and trusted traffic through vlan configuration, trunk mapping, and firewall rules.

firecrawl-build-crawl covers bulk page extraction across site sections using the /crawl endpoint, with guidance on scope constraints and when to prefer /scrape or /map instead.

hubspot skill provides cli-driven access to contacts, companies, deals, tickets, tasks, and notes via membrane framework. covers basic crud operations and search but lacks decision logic and outcome specifications.

copilotkit is a router skill that delegates to specialized sub-skills for setup, development, integrations, debugging, upgrades, and protocol work. it provides a taxonomy of tasks and references an mcp documentation server but lacks actionable procedures.

spec-driven-workflow attempts to formalize agent behavior through specification documents, but lacks concrete implementation detail and actionable procedures for practical deployment.

persona-event-coordinator bundles five google workspace skills to handle calendar creation, email invitations, and logistics tracking via predefined workflows. lacks specificity on decision logic and failure modes.

scientific-critical-thinking frames evaluation of research rigor using methodology assessment and bias detection, but lacks actionable procedural steps and omits failure modes entirely.

error-debugging-error-analysis covers systematic investigation of production incidents and root-cause analysis in distributed systems using observability tools, but lacks concrete procedural steps and decision logic.

wxt-browser-extensions catalogs 49 performance rules across service workers, content scripts, messaging, and storage. lacks actionable procedures and edge case handling, functioning as a reference index rather than executable guidance.

qiaomu-opencli-usage documents installation and basic invocation of a cli tool for automating web and electron app interactions, but lacks skill intent, decision logic, and failure handling.

axiom-apple-docs routes questions about apple framework apis, swift compiler errors, and xcode documentation to official markdown sources bundled in xcode, accessed via claude code's read tool.

deepwiki queries github repository documentation from the terminal using node.js, supporting table-of-contents extraction, full wiki retrieval, and ai-powered question-answering across single or multiple repos.

manimgl-best-practices documents patterns for 3blue1brown's opengl animation engine, covering scenes, mobjects, animations, and tex rendering through rule files and templates without executable procedures.

figma-code-connect-components matches figma design components to code implementations via code connect mapping. includes skill boundaries and prerequisites but lacks procedural detail and failure handling.

recipe-save-email-to-doc automates archival of gmail messages into google docs via sequential api calls. covers message search, retrieval, doc creation, and body insertion with gmail query parameter support.

product-hunt-launch provides a minimal cli-based workflow for generating product hunt gallery assets via inference.sh, but lacks strategic guidance on launch timing, tagline composition, maker comments, and day-of execution.

faceless-explainer converts arbitrary text into faceless video with llm-generated visuals and auto-selected style presets. includes routing logic to redirect product/site/code content to specialized skills.

infsh-cli wraps 250+ inference.sh cloud apps (image generation, video, llms, search, 3d, twitter) behind a single belt cli. installation and basic command patterns are documented, but lacks actionable skill procedures, error handling, and auth workflows.

network-engineer covers cloud and hybrid network architecture, zero-trust security, and performance optimization, but lacks actionable procedures and concrete failure mode handling.

okx-dex-trenches provides read-only on-chain research for meme token discovery and developer analysis across solana and other chains, with explicit intent classification to route write operations elsewhere.

lightpanda is a headless browser optimized for ai agents, offering performance gains over chrome for automation tasks without graphical rendering. supports javascript execution and cdp protocol.

nano-banana-pro-prompts-recommend-skill surfaces image generation prompts from a 10k+ curated library. it recommends matches based on user intent and displays sample images, with optional remix workflow for customization.

mcp-configure handles launchdarkly hosted mcp server setup during onboarding, enabling oauth-authenticated flag management tools. covers cursor and similar agents, with fallback guidance for restricted regions.

laravel documents php development principles and architectural patterns without actionable procedures. it reads as a style guide rather than a skill that executes discrete tasks.

seedance-v2 wraps ByteDance's multimodal video generator on RunComfy, accepting character images and reference audio to produce lip-synced short-form video. the skill lacks procedural depth and error handling.

react-email converts json specs into html or plain-text email output via @react-email/components, but lacks procedural depth, trigger clarity, and failure mode documentation.

scvelo estimates cell state transitions and infers developmental trajectories from unspliced/spliced mrna dynamics in single-cell rna-seq data without time-course labeling.

describe-pr generates pull request descriptions by reading repository templates and populating them with pr metadata, but lacks specificity on how descriptions are actually composed and what template sections are supported.

ai-first-engineering outlines principles for organizing teams using ai code generation, emphasizing planning rigor and architectural clarity over conventional development metrics.

muapi-seedance-2 enumerates 10 video generation modes (text-to-video, image-to-video, editing, watermark removal) with tier-specific access but lacks step-by-step procedure, input schemas, and error handling.

nansen-smart-money-alpha identifies tokens accumulating smart money inflows on solana by cross-referencing screener results against netflow labels, with optional drill-down into specific token flow intelligence.

optimization-advisor-sms prompts the user to synthesize social media performance data into an action plan, but lacks concrete procedural steps and does not define inputs, decision trees, or output contracts.

elixir-pro covers otp patterns, phoenix liverview, and concurrent systems design. delegates to external playbook for implementation details and lacks actionable step-by-step procedures.

review audits playwright test files for anti-patterns and coverage gaps. it checks for critical issues like hardcoded waits and missing assertions, then generates file-by-file scores and fix recommendations.

shannon-ai-pentester claims autonomous exploitation of web application vulnerabilities via source code analysis and concurrent exploit agents, but documentation lacks executable procedures and omits critical operational constraints.

test-skill applies red-green-refactor cycles to skill validation, requiring agents to fail without documentation before strengthening it. depends on pre-existing tdd knowledge and external examples.

anything-to-notebooklm ingests 11 content types and routes them to NotebookLM for format generation (podcasts, slides, mind maps, etc). relies on markitdown for conversion and natural language routing to 9 output modes. requires prior MCP and CLI setup.

comment-code-generate-a-tutorial refactors python scripts for beginners by applying pep 8 standards, inserting pedagogical comments, and generating markdown tutorials with setup and usage examples.

enforces no-direct-useEffect in React by requiring derived state, event handlers, and custom hooks for side effects. incomplete framework stops at rule 1 of 6.

sql-pro positions itself as a database optimization guide for cloud-native sql environments, covering query tuning and oltp/olap patterns. lacks actionable depth and specificity.

devops-iac-engineer covers infrastructure provisioning with terraform, kubernetes, and multi-cloud platforms. includes ci/cd pipeline design, observability frameworks, and security practices but lacks concrete decision trees and actionable procedures.

clerk-nextjs-skills documents clerk authentication setup for next.js app router with mcp integration, but the skill.md is incomplete and lacks actionable step-by-step procedures, decision trees, and failure handling.

eve-job-lifecycle documents phase transitions and basic job creation commands for eve, covering lifecycle states from idea through completion. lacks depth on execution patterns, error handling, and cross-job dependency management.

hr-business-partner frames strategic talent alignment across planning and performance management, but lacks concrete decision logic, API connectors, and executable workflows. mostly aspirational rather than implementable.