azure-pipelines covers local validation of azure devops yaml pipeline files using the cli, with installation steps for macos and windows. lacks guidance on actual validation workflows and failure recovery.

upstash-qstash covers serverless message queuing and scheduled task delivery via http callbacks. the skill enumerates design principles and capabilities but lacks concrete procedural guidance or trigger examples.

recent-data provides session store integration for tracking recently accessed topics, resources, and pages via initialization hooks and selector-based access patterns with built-in caching and auto-refresh.

indicator-expert provides technical analysis tooling for openalgo environments, covering real-time market data integration, 100+ indicators via numba-optimized libraries, and plotly/streamlit visualization layers. domain-locked to financial charting workflows.

weclaw-wechat-ai-bridge routes wechat messages to multiple ai backends via a go subprocess, supporting json-rpc, cli spawn, and http modes with qr login and media conversion.

3dsvg-interactive-react converts SVG files into interactive 3D React components via the 3dsvg library and visual editor, leveraging Three.js for rendering.

prose-style diagnoses sentence-level craft failures in fiction by examining diction, syntax, rhythm, and voice. it frames style as structural content rather than decoration and anchors guidance in concision and intentionality.

competitor-teardown outlines research via external search and screenshot tooling, but lacks decision logic, output schema definition, and structured analysis synthesis steps to deliver actionable competitive matrices or positioning maps as advertised.

identity-denial provides a thematic framework for constructing character arcs where protagonists resist acknowledging their own transformation, mapping denial intensity across spectrum stages. useful for writers; not actionable for systems.

anndata covers creation, storage, and manipulation of annotated biological data matrices, particularly for single-cell genomics. handles h5ad/zarr formats and integrates with the scverse ecosystem.

delegating-work teaches frameworks for identifying delegation blockers and calibrating involvement with team autonomy, drawing on product leadership practices. diagnosing obstacles, matching style to context, and designing for success are central moves.

clerk-cli is a pre-authenticated gateway to clerk's backend and platform apis, offering project management, authentication, and instance configuration. guidance is fragmented across environment constraints and incomplete procedural steps.

latex-posters covers creation of research posters for academic conferences using beamerposter, tikzposter, or baposter packages. guidance addresses layout, typography, color schemes, and multi-column designs with integration of figures and citations.

reflect consolidates session work into memory by reviewing git history, closing linked issues, and verifying phase gate compliance before declaring task completion.

generating-list-view covers creating and configuring salesforce list view metadata for filtered record listings, but provides minimal procedural depth and no concrete deployment guidance.

retro captures lightweight session observations into a learnings repository, functioning as a candidate-entry surface for move 7 of an operating loop before promotion to patterns or rules.

deepspeed covers nvme i/o operations via aio and gds handles, with blocking/non-blocking apis, pinned tensors, and parallel transfers. lacks guidance on distributed training, zero optimization, and quantization.

better-ui covers design polish through motion, surfaces, and composition details. guides reviewers on inspecting animations and references related typography and accessibility skills.

competitive-battlecard generates sales comparison documents against named competitors using web research and provided materials. outputs a templated markdown battlecard with positioning, feature matrices, objection handling, and win/loss patterns.

codegen provides tree traversal and collection utilities for extracting components, state paths, and actions from json-render ui specs. primarily import-focused with minimal procedural guidance.

guides conversion of openai apps sdk projects to mcp apps sdk by cloning reference code and applying standardized patterns for ui migration in conversational clients.

cargo-content provides CLI commands to upload, organize, and manage knowledge files (PDFs, CSVs, text) and group them into libraries for RAG-powered agent responses. moved from ai domain into separate content domain in version 1.0.19+.

tanstack-start-best-practices catalogs organizational rules for server functions, middleware, authentication, and ssr in full-stack react apps, but lacks concrete procedural steps and actionable guidance beyond naming conventions.

report generates test result summaries from playwright test artifacts. procedure is incomplete, lacks explicit output contract and outcome signal definition.

vueuse-functions guides developers to map vue.js / nuxt requirements to appropriate vueuse composables, prioritizing library functions over custom code. includes invocation rules (auto, external, explicit-only) but lacks concrete examples and decision trees.

covers video optimization for google search, sitemaps and videoobject schema. includes discovery, indexing, metadata and structured data guidance but lacks actionable procedural depth.

imaging-data-commons provides access to nci cancer imaging datasets via idc-index python bindings, supporting bulk query and download of ct, mr, and pet radiology data without authentication.

send-email wraps smtp operations with markdown-to-html conversion and template support. covers multiple providers (gmail, outlook, qq, 163/126) and file attachments, but the procedure section trails off and error handling is absent.

scientific-writing provides scaffolding for research-backed manuscript production using imrad structure, citations, and reporting guidelines. lacks concrete procedure steps, trigger conditions, and failure mode handling.

swift-formatstyle covers the FormatStyle protocol and concrete format style types for converting numbers, currencies, dates, and other values into human-readable display strings. provides type-safe composable formatting without legacy formatter subclasses.

web-quality-audit wraps google lighthouse to evaluate performance, accessibility, seo, and best practices across 150+ checks. lacks concrete procedure steps and real-world failure mode handling.

asc-id-resolver maps app store connect identifiers (apps, builds, versions) to their numeric IDs via asc cli queries, enabling downstream commands that require IDs instead of names.

postbridge-social-growth routes users through a 30-day organic growth framework for tiktok and instagram, but lacks procedural detail, decision logic, and failure mode coverage needed for consistent coaching execution.

backend-patterns documents rest api url conventions and http verb mappings. content is incomplete, lacking concrete procedures, error handling, and decision logic needed for production backend work.

academic-researcher covers scholarly writing scaffolding with emphasis on citation hygiene and source verification, outputs latex-formatted papers in ieee or apa style, but lacks concrete workflow steps and actionable triggers.

openclaw-persona-forge guides users through creating fully-realized ai agent personas (龙虾) by composing identity tensions, boundary rules, names, and avatar prompts. supports guided curation, gacha randomization, and optional image generation via pluggable skills.

cargo-analytics exports workflow run metrics, batch results, and segment data via cargo cli. covers measurement and export only, not root cause diagnosis. defers diagnostic questions to cargo-diagnostics skill.

dbs-skill-cleaner audits local agent skills for unauthorized commercial redirects, data exfiltration, and task hijacking using oecd ai principles. scans multiple vendor directories, reports findings by risk tier, then quarantines only after explicit user confirmation.

shap interprets machine learning predictions using shapley values from cooperative game theory. covers feature importance visualization, model debugging, bias detection, and production explainability across tree-based, deep learning, linear, and black-box models.

controlnet-pose routes pose-conditioned image and video generation across kling motion control, wan animate, and z-image controlnet lora. accepts openpose, dwpose, canny, or depth control inputs via runcomfy cli.

agent-evaluation addresses behavioral testing and reliability metrics for llm agents, emphasizing statistical rigor and adversarial probing over single-run benchmarks to catch production failures.

code-review-checklist provides a categorized framework for conducting code reviews spanning functionality, security, performance, and maintainability, though the procedure remains incomplete and lacks concrete decision criteria.

railway-service handles status checks, property updates, and docker-based service creation via graphql. delegates local code deployments to railway-new and github integrations to railway-environment.

career-ops-job-search wraps claude code into a job search coordinator that scores offers, generates ats pdfs, and scans 45+ company job portals. lacks explicit workflow triggers and failure recovery patterns.

caveman compresses responses by dropping articles, filler, and pleasantries while preserving technical accuracy and code. toggleable via command with intensity levels. designed for token reduction in longer conversations.

finnhub-api provides access to financial market data including stocks, forex, crypto, and news via the finnhub api. the skill covers authentication setup and basic usage patterns but lacks complete procedural detail and error handling.

prisma-compute documents cli commands for deployment and scaffolding but lacks operational guidance, failure recovery patterns, and step-by-step procedures for actual deployment workflows.