cloud provides thread and message persistence via assistant-ui sdk, supporting jwt, api key, and anonymous authentication modes with presigned file uploads.

diffity-learn is a conversational tutor that teaches technical topics through interactive projects, delegating implementation work to subagents while maintaining focus on learner interaction.

holyclaude-ai-workstation packages claude code cli, web ui, headless browser, and 5 ai clis into a single docker image, promising rapid deployment of a full coding environment without manual tooling.

eeat-signals provides a reference framework for e-e-a-t implementation (experience, expertise, authoritativeness, trustworthiness) with tactics mapped to each pillar, but lacks actionable procedures or content-specific workflows.

frontend-to-backend-requirements establishes a communication pattern where frontend developers document data needs without prescribing implementation, writing artifacts to a standardized markdown location for backend review and negotiation.

analytics-tracking guides users through setting up measurement infrastructure by first assessing business context, current state, and technical constraints. provides grounding principles but lacks concrete implementation steps.

tech-debt provides a taxonomy for surfacing and classifying technical debt across codebases. lacks actionable procedure and decision logic to move from identification to concrete remediation.

agentation bridges human visual feedback and agent code by converting UI element selection into structured css selectors and component trees. designed for iterative review loops where agents consume annotations as actionable targeting data.

sf-data handles salesforce record lifecycle operations, bulk import/export, test data generation, and cleanup. scope is clearly bounded against related skills but lacks procedural depth.

businessstrategy provides financial analysis terminology and MBA-level competency labels, but lacks actionable procedure, decision logic, or concrete output contracts.

video-processing-editing automates ffmpeg workflows for cutting, mixing, and platform-optimized exports across youtube, instagram, and social channels. lacks detailed procedures and error handling.

writing-skills documents a tdd-inspired approach to skill authoring, emphasizing test-driven validation of agent behavior against documented procedures before deployment.

qdrant-performance-optimization is a navigation hub for vector search tuning that fragments its guidance across multiple skills, covering latency, throughput, and indexing considerations without actionable procedures.

cloud-security detects iam privilege escalation, public storage exposure, and infrastructure misconfigurations through systematic posture assessment, explicitly excluding active incident response and application vulnerability scanning.

wechat-article-to-markdown fetches articles from weixin and converts them to markdown with local image storage and metadata extraction, targeting archival and knowledge base use cases.

career-transitions offers frameworks from 76 product leaders for navigating role, function, stage, or industry pivots. uses motivation diagnostics and low-risk experimentation to test hypotheses before commitment.

manim covers programmatic mathematical animation and visualization using python, suitable for creating educational video content with geometric shapes, equations, and scene-based rendering, but lacks actionable procedures.

binary-protection-bypass documents identification and bypass techniques for aslr, pie, nx, canary, relro, fortify-source, cet, and arm mte protections, with routing to related attack skills and reference to a protection matrix.

platform-metadata-api-context-get documents the salesforce metadata api's 604 types with json-structured schemas, fields, and examples. primarily a reference resource requiring programmatic section-extraction to avoid token waste rather than a procedural skill.

teach sets up a stateful learning workspace where users develop mastery across sessions through structured missions, reference materials, lessons, and learning records. lacks operational specifics for how to actually deliver or assess learning.

vulnerability-scanner documents owasp 2025 threat taxonomy, supply chain risks, and attack surface mapping concepts. lacks executable procedure and concrete trigger phrases for actual scanning workflows.

geoflow-content-automation chains ai content generation, material management, task scheduling, review workflow, and publishing into a single php/postgresql pipeline via rest api, cli, and docker compose.

defining-product-vision helps teams articulate long-term product direction using frameworks from successful leaders. covers scope clarification, future-state thinking, and specificity testing without prescribing repeatable mechanics.

treatment-plans offers latex templating for generating medical treatment documentation across specialties with emphasis on conciseness and regulatory alignment, though procedural detail and trigger definition remain underdeveloped.

rag-retrieval catalogs 30 rules across 9 categories (core rag, embeddings, contextual retrieval, hyde, agentic rag, multimodal rag, query decomposition, reranking, pgvector) but delegates actual implementation detail to external rule files. useful as taxonomy, weak as actionable skill.

incident-response outlines the incident lifecycle from triage through forensics and escalation, but lacks concrete procedures, specific triggers, and actionable decision logic needed for reliable execution.

gpt-image-2 provides text-to-image generation and editing via openai's api on runcomfy, emphasizing directive precision for multi-element prompts and embedded text over stylization.

cargo-ai covers agent creation, configuration, rag attachment, mcp server wiring, and memory management via the cargo cli. delegates orchestration, content upload, and workspace administration to sibling skills.

recipe-block-focus-time creates weekly recurring calendar blocks on google calendar to shield deep work time from scheduling conflicts, using recurrence rules to automate mon-fri protection.

mediator-pattern teaches centralized component communication via a coordinator object, contrasting direct coupling with a funnel model that handles many-to-many routing.

squirrelscan audits websites against 249+ rules across 21 categories, scoring site health and surfacing actionable fixes. covers cli installation, authentication, audit execution, reporting, and mcp integration.

pixijs-2d covers high-performance 2d rendering for sprites, particles, and interactive graphics, but lacks actionable procedures and failure mode documentation.

azure-observability exposes metrics, logs, and traces across azure monitor, application insights, and log analytics via mcp tools or cli, with kql query patterns for errors and performance analysis.

better-icons wraps iconify to search and retrieve svgs across 200+ icon libraries from a cli. installation and basic usage are documented, but the skill lacks depth on real-world usage patterns and failure scenarios.

context7-auto-research automatically retrieves current documentation for popular libraries during claude code conversations via the context7 api, with optional rate limit configuration.

vue3 maps user requests to official guide sections and api reference files, providing a file-lookup mechanism rather than actionable procedures for building vue applications.

ingesting-into-data-lake covers moving data from external sources into aws data lake tables, defaulting to s3 tables with fallback guidance for iceberg. assumes upstream connectivity exists and delegates connection setup elsewhere.

made-to-stick presents the succes framework for crafting memorable messages, but lacks concrete workflows, decision trees, or integration hooks needed for automated skill execution.

scientific-brainstorming offers conversational ideation through five phases (context, divergence, connection, evaluation, synthesis) with structured techniques like scamper and triz. lacks concrete step-by-step procedure and failure mode coverage.

product-led-sales guides teams through transitioning from pure product-led growth to sales-assisted expansion by identifying usage signals that trigger qualified leads and designing smooth handoffs between self-serve and sales motions.

launch provides strategic guidance on planning saas product launches and feature announcements, with emphasis on momentum-building and audience engagement through repeated launch cycles.

popup-generator covers popup and modal design for conversion, addressing timing, placement, and mobile compliance. guidance is prescriptive but lacks procedural depth and testable outcomes.

real-time-features covers websocket and sse patterns for bidirectional client-server communication, with reference guides and best practices checklist but incomplete implementation details.

solana covers rust and anchor framework patterns for blockchain program development, emphasizing security controls and modular code structure without concrete procedures or deployment guidance.

cold-outreach covers outbound sequence design, personalization patterns, and multi-channel cadence for prospect engagement. provides a framework (spark flow, channel mix, cadence rules) but lacks actionable step-by-step procedures or implementation guardrails.

recipe-create-classroom-course wraps gws-classroom to provision courses with customizable metadata and bulk student enrollment via email-based invitations with role assignment.

controlnet-pose routes pose-conditioned generation across kling and comfyui via runcomfy cli, but lacks procedural depth and actionable workflows for typical use cases.