guidelines for jupyter notebook workflows emphasizing pandas manipulation, matplotlib visualization, and numpy operations. covers style conventions and vectorization principles without prescriptive task sequences.

azure covers naming conventions, provisioning patterns, and operational guardrails for arm templates, bicep, pipelines, and aks. lacks procedural depth and specific trigger scenarios.

dhdna-profiler claims to extract cognitive patterns and thinking fingerprints from text using a 12-dimensional framework, but the skill.md ends abruptly before detailing the actual procedure, decision logic, or output structure.

creates admin users in medusa via npx command, accepting email and password arguments and reporting success or failure with next steps.

gog wraps google workspace apis (gmail, calendar, drive, sheets, docs, contacts, tasks) via cli with support for search, file operations, event management, and data export. documentation is fragmented and incomplete.

context7 retrieves current library documentation via api query, useful for augmenting stale training data with live version-specific guidance on frameworks and components.

capacity-planning provides overhead-adjusted capacity calculation formulas and best-practice checklists for team resource allocation, but lacks executable procedures and concrete decision logic.

recipe-transfer-file-ownership automates bulk ownership reassignment of google drive files across users via the gws-drive api, requiring two sequential cli operations with an irreversible consequence.

fusion-help-docs covers authoring and publishing help content via the fhelp CLI, but lacks actionable procedures, failure handling, and implementation detail beyond listing use cases.

avatar-video wraps heygen api endpoints for creating animated videos from avatars or images, supporting voice synthesis and script control via two discriminated union types.

infographics wraps nano banana pro ai for automated visual generation with gemini 3 pro quality gates and optional perplexity sonar research integration, targeting users without design skills.

compression covers javascript payload reduction via gzip and brotli. lacks concrete step-by-step procedures, decision logic, and failure mode discussion. positioning is general but context-appropriate.

runpod enables serverless gpu inference via pay-per-second cloud processing. covers account setup and docker image deployment but lacks operational guidance for endpoint management, scaling, or troubleshooting.

llm-icon-finder locates and constructs cdn urls for ai model and provider brand icons from the lobe-icons github library, supporting svg, png, and webp formats across light and dark theme variants.

virtual-lists covers windowing techniques to render only visible rows in large datasets, improving performance for lists and grids with hundreds or thousands of items.

alicloud-ai-video-wan-video-test is a minimal smoke test for wan video generation that validates the happy path, records outcomes, and requires manual setup without automation or recovery guidance.

sentiment-analysis provides working python code for lexicon and machine learning approaches on sample data, but lacks actionable intent, real input contracts, and failure handling. suitable as educational reference, not production skill.

torch-geometric covers graph neural network construction and training with pyg's standard layer library, data structures, and mini-batch support, but lacks complete procedural depth and failure guidance.

zai-tts wraps glm-tts text-to-speech generation with support for three built-in voices and custom clones, offering speed and volume control via command-line parameters.

conductor-manage covers track lifecycle operations (archive, restore, delete, rename, cleanup) with confirmation gates for destructive actions and backup recommendations, but lacks concrete step sequences and detailed failure recovery paths.

terraform-infrastructure documents terraform fundamentals and multi-cloud provisioning patterns, but lacks concrete decision logic, error handling, and actionable procedural steps beyond setup.

revealjs covers html-based presentation creation with reveal.js, focusing on design-first thinking before implementation. lacks concrete procedural steps and omits handling of common constraints like browser compatibility or large slide decks.

statsmodels covers regression, glm, time series, and hypothesis testing for statistical inference. lacks procedural depth and specific trigger patterns for cross-vendor integration.

init generates or refreshes AGENTS.md guidance files by filtering for non-discoverable workflow conventions and gotchas, using a discoverability check as its primary decision gate.

oauth-integrations documents github and microsoft entra oauth flows for edge runtimes like cloudflare workers, covering provider-specific quirks, token lifetime differences, and common configuration pitfalls.

prompt-caching covers caching strategies for LLM cost reduction, including anthropic prompt caching, response caching, and cache augmented generation patterns with identified anti-patterns and sharp edges.

geo-database documents ncbi geo's data hierarchy and use cases for gene expression research, but lacks concrete procedural steps, api endpoints, or executable workflows for actual dataset retrieval.

image-enhancer upscales and sharpens screenshots and photos for presentations, documentation, and social media. it analyzes resolution and compression artifacts, then applies enhancement filters and outputs optimized files while preserving originals.

biopython is a python toolkit for molecular biology tasks including sequence manipulation, file parsing, and database access. the provided skill documentation is incomplete and does not specify concrete trigger patterns, procedures, decision logic, or error handling.

filesystem-context offloads agent memory and state to files as a context overflow mechanism, supporting dynamic retrieval of shared state across agent trajectories without bloating token consumption.

confluence-expert provides basic mcp operations for space and page management but lacks actionable workflows, decision logic, and guidance on permission hierarchies or failure scenarios.

v3-security-overhaul coordinates threat modeling, cve remediation, and security testing via delegated agents. lacks concrete step-by-step execution paths and operational clarity.

configuration-management covers environment setup, secrets handling, and 12-factor principles across environments. provides best-practice checklists but lacks executable procedures and failure-mode guidance.

eastmoney-select-stock queries the eastmoney api via natural language to filter stocks across a/h/us markets and export results as csv, requires api key setup.

agent-tdd-london-swarm documents london school mockist tdd patterns with code examples, but lacks actionable procedure for actual swarm invocation and omits failure handling.

create-component generates adobe aem components with configuration validation. intent is clear but procedure lacks detail, edge cases are minimally addressed, and trigger phrases are absent.

implementing-llms-litgpt documents model loading and basic setup using litgpt's pretrained architectures, but lacks training workflows, validation procedures, and failure handling details.

tensorrt-llm optimizes llm inference on nvidia gpus with high throughput and low latency. covers quantized model deployment, multi-gpu scaling, and real-time serving use cases. comparison to vllm included.

latex-posters provides high-level context for creating conference and academic research posters using beamerposter, tikzposter, or baposter, but lacks procedural depth and concrete workflow steps for execution.

risk-management-trading articulates position-sizing philosophy and drawdown mechanics through trader anecdotes and principles, but lacks procedural steps, trigger phrases, and concrete decision logic for actual implementation.

html-ppt-skill provides static html/css/js presentation templates with 36 themes and 31 layouts, requiring no build tooling. actual procedural depth and failure handling are absent from the documentation.

synthetic-monitoring provides a playwright-based framework for simulating user journeys and api flows, with best practices guidance but incomplete procedural depth and missing deployment context.

reader-recap surfaces highlights and notes from readwise reader over a configurable time window, with optional persona-aware contextualization for personalized briefing tone.

trade-hypothesis-ideator generates ranked hypothesis cards from normalized input bundles, applies critique filtering, and exports viable hypotheses to strategy artifacts. domain-specific to trading/investment thesis development.

security-compliance-compliance-check provides a compliance audit framework covering gdpr, hipaa, soc2, and pci-dss. lacks concrete procedures and fails to specify actionable steps, decision logic, or verifiable outcomes.

shopify covers theme development fundamentals including liquid templating, online store 2.0 architecture, and performance optimization, but lacks actionable procedures and failure-mode guidance for real implementation scenarios.

add-app-to-server covers pairing existing mcp tools with html uis via the apps sdk, allowing inline rendering while preserving text-only fallback compatibility.