testing-agentforce covers automated test execution for agentforce agents including smoke testing, batch runs, and trace analysis, but the specification is incomplete and lacks procedural clarity in critical areas.

asc-aso-audit performs offline validation of app store metadata against aso rules, then identifies keyword gaps via astro mcp. requires pre-pulled canonical metadata and optional astro tracking.

marketing-loops guides setup of recurring AI-driven marketing workflows with defined triggers, bounded steps, and stopping conditions. establishes cadence-based execution for tasks like seo scanning and churn monitoring, wiring together other marketing skills.

chrome-devtools provides browser automation and debugging via MCP, offering page navigation, snapshot inspection, and element interaction through persistent Chrome sessions with configurable extensions and memory tooling.

aws-cdk-python-setup covers baseline installation and project initialization for aws cdk development in python, including credential configuration, virtual environment setup, and core deployment commands.

tag-taxonomy enforces consistent wiki tagging by normalizing against a controlled vocabulary file. it relies on config resolution and a canonical taxonomy definition but lacks concrete procedural steps and failure handling.

pestel-analysis provides a macro-environmental scanning framework for identifying external threats and opportunities across six dimensions. intended for strategic planning timelines of 12-24 months, it covers political, economic, social, technological, environmental, and legal forces that shape product direction.

data-scientist role covers ml lifecycle stages from problem definition through evaluation and communication, but lacks concrete input specs, decision logic, and failure handling needed for agent execution.

ui-audit provides a decision framework and reference library for evaluating interfaces against ux principles, grounded in jobs-to-be-done and macro business bets. the skill documents audit output schema and checklist coverage but lacks concrete trigger phrases and failure mode guidance.

executes local diff checks simulating github pr validation workflow. routes changes to sub-agents that read artifact files directly, avoiding context bloat.

video-download wraps yt-dlp to fetch videos, audio, and subtitles from urls at configurable resolutions. provides command templates for common tasks but lacks structured intent/input/output contracts.

docs-architect generates long-form technical documentation by analyzing codebases for architecture, design patterns, and component interactions. produces structured markdown with diagrams, rationale, and multi-audience reading paths.

company-research uses browserbase search and a plan-research-synthesize pattern to generate scored research reports and csvs for target companies, requiring api key setup and cli approval.

autonomous-agent-patterns outlines design patterns for autonomous coding agents including think-decide-act loops, tool integration with schema definitions, permission systems across four approval levels, and sandboxing strategies. covers checkpoint/resume mechanics and mcp integration for dynamic tool discovery.

hybrid-cloud-networking covers connectivity patterns and configuration snippets for aws, azure, gcp, and oci. lacks procedural depth, decision trees, and explicit failure recovery paths.

vaex handles out-of-core dataframe operations on billion-row datasets, enabling lazy evaluation and fast aggregations without requiring data to fit in ram. covers format conversion and interactive visualization.

momentic-test documents the mechanics of Momentic's AI-driven E2E testing framework: step execution, caching behavior, timing, file formats, JavaScript contexts, and MCP authoring workflows. Comprehensive reference for building and maintaining YAML test files.

step-parts retrieves real cad geometry for off-the-shelf electronics and mechanical components from a hosted api, prioritizing exact matches over placeholder substitutes with documented fallback behavior.

mongodb-mcp-setup guides users through configuring the mongodb mcp server with three authentication options: connection string, atlas admin api credentials, or x.509 certificates. covers environment variable setup and method selection.

auth-web-cloudbase covers frontend integration for cloudbase web authentication across 8 login methods with user management, session handling, and token retrieval for api calls.

codex-cli-runtime wraps the codex-companion task runner for the rescue subagent, enforcing a single forwarding pattern with optional prompt rewriting via gpt-5-4-prompting before delegation.

sector-analyst ranks sectors and estimates market cycle phase by analyzing uptrend ratios from a public csv dataset, with optional chart image support to contextualize risk regime positioning.

workflow-orchestration-patterns documents when to apply temporal-based orchestration in distributed systems, emphasizing use cases over concrete implementation. introduces saga patterns and state management but lacks procedural depth.

contact-page-generator covers contact form design, field optimization, and page layout for conversion. it references best practices like field count, mobile responsiveness, and trust signals but lacks procedural depth and failure handling.

msgraph provides offline access to 27,700+ microsoft graph apis via indexed search and optional execution, covering endpoints, permissions, schemas and samples without network overhead.

writing-web-documentation covers principles for creating developer-facing docs that prioritize fast onboarding, clear information architecture, and scannability over product-fact dumps.

dynamic-import covers lazy-loading modules in react via suspense and react.lazy to defer non-critical component initialization, reducing initial bundle size for faster page loads.

astropy provides coordinate transformations, fits file handling, unit conversions, and tabular data operations for astronomical research. covers cosmological calculations and time scale conversions but lacks procedural depth.

robots-txt handles configuration and auditing of robots.txt files to control search engine and ai crawler access at the path level, with clear scope boundaries around crawl vs index control.

bnbchain-mcp covers connecting to bnb chain via model context protocol and invoking tools for blocks, transactions, smart contracts, tokens, nfts, wallets, and greenfield via private key authentication or read-only modes.

azure-diagrams converts arm templates, cli output, or descriptions into eraser cloud architecture diagrams. parsing is competent but procedure lacks numbered steps and the skill omits error handling for malformed input or api failures.

eve-skill-distillation teaches how to recognize repeated work patterns and encode them as reusable skills using an orchestrator pattern for multi-skill captures. targets teams looking to reduce cognitive load and knowledge silos across jobs.

temporal-python-testing covers pytest-based testing strategies for temporal workflows, including unit testing with time-skipping, integration testing with mocked activities, replay testing for determinism validation, and ci/cd pipeline setup.

codeql covers workflow setup and cli usage for static analysis via github actions and standalone scanning. lacks procedural depth on common failure modes and remediation paths.

django-celery-expert routes requests across asynchronous task design, broker/backend configuration, error handling strategies, and production monitoring patterns. lacks step-by-step procedures and concrete implementation examples.

chat-sdk provides a unified typescript interface for authoring bots once and deploying across eight chat platforms. developers get normalized event handling, ai sdk streaming, and jsx-based interactive cards, with pluggable state and custom adapter support.

generates reference documentation for library public surfaces (components, functions, commands) by reading source code directly, with checklist tracking for single or batch entries. focuses on extracting type signatures and defaults rather than usage patterns.

backtesting skill generates complete vectorbt scripts with multi-strategy support, TA-Lib indicators, and NIFTY benchmark comparison, but lacks explicit execution flow and error handling details.

sentry-nestjs-sdk covers trigger conditions and version scope for nestjs error monitoring setup, but lacks actionable step-by-step procedure, decision logic, and explicit output validation.

ai-prompt-engineering-safety-review provides a multi-dimensional framework for analyzing ai prompts across safety, bias, and security concerns. it delivers risk scoring and improved prompt versions but lacks concrete procedural steps and actionable trigger conditions.

provider-actions covers terraform plugin framework implementation for imperative lifecycle operations, with experimental before/after create and update triggers, polling patterns for async workflows, and error handling for long-running tasks.

testing-agentforce automates validation of agentforce agents through smoke tests, batch execution, and trace analysis. covers iterative fix loops but procedure remains incomplete in the provided skill.md excerpt.

antigravity-workflows orchestrates multi-step task sequences across distinct domains (saas, security, ai agents, qa) but lacks procedural detail and failure handling to execute reliably without human intervention.

dogfood guides exploratory qa testing on web apps via browser automation, capturing evidence and producing bug reports with limited procedural specificity.

qa guides users through conversational bug reporting, explores relevant code sections in parallel, and translates findings into github issues using domain language. incomplete spec.

metal-kernel covers implementing native Metal kernels for PyTorch operators on Apple Silicon via c10/metal infrastructure rather than MPSGraph, with workflows for new operators and MPSGraph migration.

aws-sdk-java-v2-secrets-manager covers retrieval, deserialization, and Spring Boot integration patterns for Secrets Manager, but lacks explicit procedures and failure handling details.