doc generates and validates project documentation by detecting project type and scanning for existing docs, but lacks complete workflow definition and recovery mechanisms.

finance-news aggregates multi-region market data via rss and apis, generates summarized briefings with language options, and delivers via whatsapp with cron scheduling and portfolio tracking support.

lark-okr provides shortcuts for managing飞书 OKR cycles, objectives, key results, and progress records, but lacks detailed procedures and error handling guidance.

specs-code-cleanup removes debug logs, dead code, and temporary comments from reviewed tasks to prepare code for production. focuses on cosmetic improvements within a specification-driven workflow but lacks procedural detail.

quality-manager-qms-iso13485 outlines iso 13485 qms workflows for medical device orgs, but lacks actionable procedure depth, specific trigger phrases, and failure mode handling needed for operational use.

ghost-report aggregates findings from multiple security scanners into a single prioritized report, with caching at commit level and severity-based sorting, but lacks clear procedural steps and concrete trigger patterns.

aws-sdk-js-v3-usage documents package structure and import conventions for the aws sdk javascript v3, covering client organization and module resolution patterns.

partial guide to crewai agent design emphasizing task-first thinking and agent minimalism, but lacks actionable procedures for role/goal/backstory definition, llm selection, or tool assignment.

tushare-data maps chinese natural language financial queries into executable data workflows using the tushare api, covering stock trends, valuation lookups, comparative analysis, fund flows, and macro indicators with csv export.

ux-researcher-designer outlines four workflows for persona generation, journey mapping, usability testing, and research synthesis, but lacks actionable procedural detail and failure mode acknowledgment.

seo-image-gen wraps gemini image generation for og/social preview and blog hero images, mapping seo use cases to optimized aspect ratios and resolution, but lacks complete workflow documentation and error handling.

scholar-evaluation provides a generic academic review framework without concrete evaluation criteria, scoring rubrics, or actionable decision logic. intended for paper and proposal assessment but lacks specificity needed for consistent application.

video-extend chains or extends video clips using google veo 3-1 endpoints via the runcomfy cli, supporting narrative shot-by-shot generation from a seed prompt.

yt-dlp-downloader wraps the yt-dlp command-line tool to fetch videos from youtube, bilibili, twitter and many other platforms. covers dependency checking but lacks actionable download procedures and failure handling.

a micro-prompt reinforcing interactive programming workflows that reminds agents to treat the live repl as source of truth, communicate evaluated code clearly, and maintain todo lists during development.

fact-check provides conceptual framework for verification-as-separate-pass but lacks concrete procedure, trigger definitions, and failure handling. remains largely theoretical.

health audits agent config alignment across instruction surfaces, tools, and verifiers. it flags mismatches between declared behavior and runtime state, with budget-aware escalation to deep inspection when drift is suspected.

kibana-streams covers read-only stream metadata, settings, queries, and lifecycle operations (enable, disable, resync) via the experimental kibana streams rest api. mutating operations not yet supported.

marketingskills-ai-agents packages marketing frameworks (aida, pas, jtbd) as markdown skills for coding agents, but lacks concrete procedures, specific trigger phrases, and failure-mode guidance needed for reliable execution.

convex-component-authoring covers structuring isolated, reusable convex packages with type-safe exports, react hooks, and npm publishing, but lacks actionable step-by-step guidance on actual implementation.

ipados-design-guidelines covers apple's ui patterns for ipad-native apps, including adaptive layouts, multitasking, pointer support, and keyboard integration. incomplete and truncated.

algo exposes binance futures and spot algo trading endpoints (twap, volume participation) via authenticated api calls, returning json responses with no procedural guidance on execution flow.

setup installs the parallel cli tool across multiple platforms and configures authentication, with fallback path configuration for discovery issues.

wecomcli-smartsheet manages enterprise wechat table structures and records via cli commands, supporting document location by id or url. lacks procedural depth and failure mode handling.

debugging provides trigger phrases for common failure modes and begins a framework for systematic troubleshooting, but lacks procedural depth and decision logic beyond initial data collection.

turborepo-caching covers monorepo build optimization through local and remote caching configuration, but stops short of actionable implementation details and concrete configuration examples needed for real-world deployment.

weatherkit wraps apple's weather service to retrieve forecasts, alerts, and historical data across swift/ios platforms. requires explicit attribution display and targets ios 26+.

recipe-review-meet-participants retrieves attendance records and session duration from google meet conferences via three sequential api calls, extracting participant and session metadata.

printing-press-score evaluates generated CLIs against the steinberger bar with support for rescoring, named lookups, path resolution, and side-by-side comparison. documentation cuts off mid-setup.

gke-basics provides minimal scaffolding for gke cluster creation via gcloud and kubectl commands, assumes user familiarity with kubernetes and google cloud, references a golden path document without inline specification.

lark-vc-agent enables bot-mediated actions during active lark video conferences, including joining, leaving, and reading realtime meeting events. currently in closed beta with limited user access.

clawhub provides cli commands for skill discovery, installation with version pinning, intelligent hash-based updates, and registry publishing with authentication support.

unreal-engine-cpp-pro outlines when to apply ue5 c++ best practices but lacks concrete procedural steps, decision logic, or failure recovery guidance needed for expert-level work.

azure-compliance wraps azqr scanning and key vault expiration checks, classifying findings by priority but lacks concrete procedures, error handling, and specific trigger phrase patterns.

pragmatic-programmer teaches meta-principles from hunt & thomas for evaluating software craftsmanship: dry, orthogonality, tracer bullets, design by contract. best suited for architecture reviews and culture guidance, not tactical execution.

ai-avatar-video routes text and audio to multiple runcomfy avatar models (omnihuman, wan, happyhorse, seedance) with lip-sync, returning invoke commands and prompts for the selected path.

design-sprint frames a five-day process for de-risking product decisions through prototyping and user testing. lacks detailed execution steps, decision criteria, and practical recovery paths for common obstacles.

planning-with-files-ar outlines a manus-style file-based task tracking system using markdown documents (task_plan.md, findings.md, progress.md) to maintain persistent work context. core flow begins with context recovery from existing files and session state, but lacks specificity on outputs and trigger conditions.

insforge-debug provides observability primitives and symptom recipes for diagnosing backend issues, but lacks concrete procedures and complete decision trees for the debugging workflow.

recipe-bulk-download-folder automates batch retrieval of files from a google drive folder, with optional format conversion for docs and sheets to pdf or other formats during download.

relight adjusts lighting properties on static images by routing requests to specialized endpoints, supporting both dedicated LoRA models and identity-preserving edits via runcomfy cli.

capture-api-response-test-fixture provides conventions and helper patterns for storing provider api responses as test fixtures, with separate approaches for generatetext and streamtext patterns.

test-cases generates structured test scenarios from requirements documents, mapping functional and edge case coverage. lacks concrete procedural steps and demonstrates incomplete structural definition.

sf-metadata covers salesforce schema inspection and metadata artifact generation via sf cli, with clear ownership boundaries against related deploy and permissions skills but no actionable procedure.

prompt-master generates deployment-ready prompts for specific ai tools by extracting user intent and optimizing token usage. operates only when explicitly requested for prompt generation, improvement, or adaptation.

postgresql-code-review demonstrates postgres-specific pattern matching on jsonb, array, and indexing concerns but lacks systematic procedure, clear decision logic, and explicit failure handling.

storybook-story-writing describes csf3 format requirements and component variation patterns but lacks a structured procedure for implementing stories from first principles. documentation is competent but procedure-level guidance is vague.