box is a connector skill that bridges ai agents to box cloud storage via the membrane cli, handling oauth authentication and api discovery, but lacks concrete workflow examples and actionable decision logic.

experience-map describes a conceptual framework for visualizing user journeys across multiple channels and phases, but lacks actionable procedures and concrete decision criteria for practitioners to execute the mapping work.

mem0 provides a managed memory layer for ai agents. the skill documents basic client installation but lacks complete operational procedures for storing, retrieving, and managing agent memories.

pgvector-semantic-search covers storing and querying embedding vectors in postgresql with hnsw indexing for approximate nearest neighbor lookups. provides golden path configuration for cosine distance queries at scale.

firebase-firestore covers nosql document storage with real-time sync and offline caching. lists capabilities like compound queries and batch operations but lacks actionable procedures for implementation.

convert-plaintext-to-md transforms plain text documents to markdown using explicit instructions, reference templates, or documented options. supports multiple markdown flavors and includes formatting cleanup utilities.

effect-index routes developers to appropriate Effect skills via a decision tree covering 12 domains (concurrency, streams, DI, HTTP, validation, testing). lacks actionable procedures and failure mode awareness.

gpt-image-2 wraps chatgpt's image generation (text-to-image, editing, style transfer) through a local codex cli, requiring an active chatgpt plus/pro subscription and no separate api keys.

plannotator-visual-explainer routes content into themed HTML outputs via three conditional paths (plan, pr explainer, catch-all visual), with delivery constrained to plannotator's annotation ui.

longbridge-quant provides trigger conditions and data-source policy for quantitative strategy discussion, but lacks step-by-step procedures, failure handling, and actionable workflow definition.

automate-github-issues outlines a five-phase pipeline for dispatching parallel jules agents against open github issues, but lacks concrete implementation steps, trigger examples, and failure handling.

ethereum covers solidity best practices, openZeppelin patterns, and hardhat tooling for smart contract development. prescriptive on code standards and security patterns but lacks procedural depth for executing specific tasks.

skills-search delegates to ccpm CLI for registry operations. the file specifies execution behavior and bootstrap logic but lacks concrete search/discovery use cases and outcome signals.

compound-pattern describes a component architecture for building related ui elements that share state, like dropdowns and tabs, but lacks concrete implementation guidance and specific examples.

product-hunt-launch provides a CLI workflow for generating gallery images and optimizing launch assets via inference.sh, but lacks decision logic, timing strategy, and launch-day tactics described in the slug.

surf wraps a crypto data cli with 83+ commands across 15 domains, but the skill.md is a partial setup fragment lacking intent definitions, trigger phrases, decision logic, and outcome signals.

react-grab copies react component metadata (name, path, html) to clipboard on hotkey, enabling ai agents to inspect ui elements in browser for code generation tasks.

llm-application-dev covers structured prompt engineering patterns and basic llm interaction setup without detailed rag integration, context management, or failure scenarios. incomplete skill definition with minimal actionable procedure.

agent-browser wraps playwright for web automation via inference.sh, exposing element interaction through @e refs and screenshot/video capture. incomplete documentation leaves implementation details and capability boundaries unclear.

resend-design-skills is a routing index that directs to four downstream design resources (brand guidelines, design system, marketing patterns, design audit) without prescriptive steps or actionable procedures for any individual task.

nextjs-typescript-tailwindcss-supabase documents architectural preferences and coding conventions for next.js 14 full-stack development, but lacks actionable procedures or concrete decision triggers.

seedance-v2 wraps ByteDance Seedance 2.0 Pro for multimodal cinematic video generation on RunComfy, supporting character images, scene video, and audio inputs with native lip-sync, but lacks actionable procedure detail.

framework selection guide for agentic ai development, comparing pydantic ai (python) and claude sdk (node.js) with minimal procedural depth.

java-maven covers pom configuration, dependency management, and multi-module project setup. lacks detailed procedures and executable steps, treating maven as a reference rather than a guided workflow.

ministack-aws-emulator provides local aws service simulation across 25+ services including s3, sqs, dynamodb, and lambda, running on a single configurable port without credentials or external dependencies.

clawdirect indexes social web experiences for ai agents via a browsable directory. supports liking entries and adding new sites through atxp-based mcp tools, but lacks procedural detail and error handling.

apollo-client covers setup and integration patterns across react, next.js, and tanstack environments, but lacks concrete procedural depth and actionable query/mutation examples.

anything-analyzer-cdp is an electron desktop app that intercepts browser network traffic via chrome devtools protocol and sends captured data to ai models for protocol analysis report generation.

gpt-image-1-5 wraps openai's image generation and inpainting apis with cli tooling. supports variable quality, size, and background modes but lacks explicit error handling and incomplete trigger documentation.

signup-flow-cro provides a broad reference library of signup optimization patterns, principles, and experiment ideas but lacks concrete procedural steps or decision logic to guide users through actual optimization work.

cudf acceleration guidance for gpu dataframes, addressing compatibility matrices, library wording conventions, and implementer decision paths between cudf.pandas and explicit cudf migrations.

expo-router covers file-based routing, dynamic routes, groups, and navigation patterns. reference-heavy structure with incomplete actionable guidance and minimal failure mode documentation.

pexo-agent handles full-pipeline video generation from text/image/audio with auto model selection across 10+ vendors and multi-turn dialogue, but lacks actionable step-by-step procedure and concrete failure handling.

do-competitively orchestrates multi-agent task execution via competitive generation, meta-judge evaluation rubric synthesis, and evidence-based output merging. strategy selection adapts to confidence levels and cost constraints.

gen-paylink-govilo packages files into paid unlock links via govilo api, collecting crypto payments with minimal setup. covers file upload and link generation but lacks error handling and complete procedural detail.

swift-style documents naming, organization, and syntax conventions for swift code. it covers golden path logic, access control, memory management, and common mistakes but lacks executable procedures or agent decision logic.

readwise-mcp exposes highlights and reader documents via mcp protocol. setup requires http configuration pointing to readwise endpoint. lacks actionable procedures for common tasks.

dingtalk-ai-table wraps mcporter CLI to expose base, table, field, and record operations against dingtalk's ai table service via baseId/tableId/fieldId/recordId identifiers. covers crud at the record level with minimal scaffolding documentation.

free-code-claude-cli wraps a telemetry-stripped, guardrail-removed fork of claude code cli with 54 unlocked experimental features, supporting five model providers.

diffdock predicts 3d protein-ligand binding poses using diffusion models, suitable for structure-based drug discovery. supports pdb structures, multiple ligand formats, and confidence scoring but requires external scoring for binding affinity estimation.

remotion-render converts tsx component definitions into mp4 video files using the remotion library, executed through inference.sh. flow requires belt cli authentication and handles remotion api surface (frame hooks, video config, animation primitives).

seo-sxo analyzes search result patterns to identify page-type mismatches and alignment gaps between content strategy and search intent signals, using serp structure as a diagnostic tool for ranking potential.

logo-generator collects brand metadata through conversational intake and promises svg and presentation output, but lacks concrete generation logic, output guarantees, or failure handling.

twitter-automation wraps nine twitter/x cli commands for posting, engagement, and account management. media support and inference.sh integration enable content workflows, but documentation lacks depth beyond basic examples.

polyglot-integration documents strategies for calling code across language boundaries (node native addons, python, rust, grpc, java) with incomplete procedures and vague decision criteria.

flutter-drift provides basic guidance for adding or repairing drift persistence in flutter apps, using sqlite and type-safe dart queries, but lacks detailed procedure steps and failure mode coverage.

feedback-page-generator guides creation of feedback and roadmap pages that collect user input via integrations like canny, featurebase, or uservice, targeting product-led growth and community engagement.