natural-language-reminder parses time expressions and schedules one-time or recurring cron jobs, logging activity to markdown. covers relative, absolute, and date-based time formats with both interval and calendar-based recurrence.

vue covers composition api patterns, component architecture, and testing for vue 3 projects. routes queries to specialized reference files based on task context to minimize token overhead.

notfair-meta-ads-agent audits and manages meta ads accounts across facebook and instagram via hosted mcp, supporting read-only diagnostics and approval-gated writes for campaigns, budgets, and creatives.

deskpeto converts photos into animated desktop pet packages for chatgpt/codex, guides installation, enables in-chat pet summoning, and offers free community pets from a shared gallery.

infographic skill provides structured layout templates and design specifications across five infographic types, guiding users through goal definition and content organization. output format is prescriptive but requires external designer implementation.

oo-docuseal wraps the docuseal api via oomol cli tooling, enabling template retrieval, submission creation, and listing without direct token handling. three read and write actions available.

custom-telegram-bot enables agents to operate branded telegram bots for two-way messaging, file sharing, and inbound message handling through agentpmt-hosted remote calls. supports text, photos, documents, and cursor-based update polling.

remotejobhunter automates remote job discovery by searching multiple sources, matching against user profile, and delivering reports via email or console. supports daily scheduling and link verification.

client-project-manager tracks freelance work across clients, projects, time, and invoices with structured json storage. covers basic crud operations, status dashboards, invoice generation, and weekly client updates.

investment-risk-scanner applies buffett's five-layer framework plus porter's five forces to evaluate business quality, moat sustainability, and hidden leverage (especially supply chain financing). includes ocf quality analysis and case studies (byd, rivn, tesla, etc) but procedure lacks step-by-step execution flow.

incident-responder provides three structured playbooks (data exfiltration, ransomware, credential compromise) with concrete bash command sequences for triage, evidence collection, and containment. playbooks lack decision logic and failure recovery; documentation reads as template-heavy without actionable guidance on interpreting results.

babysit-pr polls a github pull request for review comments, ci status, and mergeability, continuing until merge, closure, or a situation requiring intervention arises.

insurance-agent-trainer offers a training framework for insurance agents with product parsing, question generation, and role-play simulation. the architecture is conceptual, treating all procedures as educational templates. compliance disclaimers are extensive, but execution clarity and edge-case handling are weak.

bili-tech-video-automation provides a six-stage workflow for video content product management, from requirements planning through post-launch review, using EARS syntax for requirements documentation. targets video translation, subtitle automation, and cross-platform distribution saas products.

notion-api-integration covers debugging and building Notion API work: authentication, database schema, page and block operations, filtering, pagination, bulk migrations, webhooks, file handling, and common failure modes with recovery paths.

hazardous-chemical-id identifies whether chemicals meet china's regulatory definitions across seven classification schemes, outputting structured reports with hazard categories, monitoring status, and critical quantities per gb 18218-2018.

travel tracks a standing system for passports, visas, day counts, bookings, points and budgets across multiple trips. it answers entry questions before dates are fixed, records reservation deadlines, handles disruptions, and closes loops with debriefs. not for building single-trip itineraries or fare search.

codexbox runs openai codex behind http rest, openai-compatible chat, mcp, telegram and cron interfaces with bearer-token auth per surface. use it to drive codex programmatically instead of only in a terminal.

nontender-skill-1-0-0 assists chinese state enterprises in determining non-tender procurement methods (competitive bidding, price competition, negotiation, direct purchase) against regulatory boundaries and internal policy, with explicit disclaimers that it supports rather than replaces procurement decision authority.

money advises on debt sequencing, emergency-fund sizing, affordability, and allocation priorities. it works from jurisdiction-aware defaults, writes decisions to persistent files, and applies real after-tax rates to all comparisons.

tailwindcss debugs utility application, theme configuration, build integration, and framework wiring. covers scanner mechanics, cascade conflicts, dark mode, variants, and component APIs across v3 and v4, with situation-indexed guides and user-preference memory.

fastmail-api covers jmap automation for mail search, triage, sending, masked addresses, contacts and calendars. includes session discovery, bulk operations with state tracking, and detailed failure signatures. procedural depth varies by operation.

openclawlog provides wordpress blog management via xml-rpc and rest apis, covering user registration, post and page creation, media uploads, taxonomies, and comments. heavily code-example driven but lacks structured intents and error recovery guidance.

agent-trading-atlas provides a collective decision-sharing layer for trading agents, enabling outcome tracking and wisdom queries against cohort performance. agents retain their own analysis tools while accessing shared experience records and decision evaluation.

claw-mission-killer stops running agents and rolls back their transcript to the state before the triggering message. supports fuzzy agent name matching and backs up interrupted sessions for recovery.

economic-damages-calculation-memo guides forensic accountants through structured intake, loss-period anchoring, method selection, incremental-cost analysis, mitigation, time-value treatment, tax handling, and sensitivity scenarios to produce defensible draft calculation memos for litigation damages (lost profits, ip infringement, wrongful termination, personal injury) with explicit source tracing and assumptions logging.

wechat-aiagent-dev guides merchants through structuring ecommerce and service capabilities for wechat mini-program ai integration, offering mode selection advice, data architecture templates, and code generation for both development and auto modes.

upkuajing-global-company-people-search-zh queries global enterprise and contact databases via five python scripts (two list searches, three detail/contact endpoints), supports filtering by product, industry, role, country, and returns company registration data plus contact channels including email, phone, whatsapp.

gingiris-b2b-growth covers plg vs slg motion selection, five-stage revenue scaling paths from zero to $10m arr, pricing discipline, and 2b deal execution including payment structures and scope control. multi-language reference for b2b saas operators.

affaan-m-configure-ecc is an interactive installer that walks users through selecting, installing, and verifying skills and rules for the everything claude code project across user or project-level directories, with optional post-install optimization.

ce-strategy guides creation and maintenance of STRATEGY.md, a structured product anchor capturing problem, persona, success metrics, and investment priorities. downstream skills depend on it for grounding.

push-notifications covers local and remote notification implementation on ios/macos, spanning permission flows, apns registration, payload structure, foreground handling, actions, grouping, and rich notifications with ios 16+ support.

minsu-operator covers end-to-end guesthouse operations including site assessment, design, dynamic pricing, ota platform management, customer service workflows, and financial metrics. triggers on chinese hospitality keywords and provides scored evaluation models and reference tables but lacks actionable recovery steps for common failures.

agentic-workflow-designer maps business processes for ai automation, scores suitability, generates platform-specific workflow specs (n8n/make/zapier/langgraph), and calculates roi. covers platform selection and human-in-the-loop design.

rollinggo-hotel-booking-global covers hotel search, filtering, room comparison, and booking across 2 million hotels via cli commands. requires login auth, two-step confirmation before payment, and price-locking via referenceNo. supports 200+ countries with credit card payment and 24/7 support.

text-from-video extracts spoken text from uploaded video files (mp4, mov, avi, webm, up to 500mb) via remote gpu processing. users describe desired output and receive transcripts or edited mp4s within 30-90 seconds, with support for batch operations and iterative refinement.

breakcold-crm covers six headline workflows (auto-tasks, pipeline movement, reports, prospect research, inbox contact detection, crm setup) plus universal plumbing for speed and error-proofing. operates through the breakcold mcp server with caching, batch processing, and breadcrumb trails.

free-maker-image converts still photos into 1080p video via remote gpu rendering, handling auth, session management, sse streaming, and export workflows with explicit api contracts and error codes.

lh-video-gen transforms markdown scripts into vertical 9:16 mp4 videos by parsing sections, synthesizing tts audio, rendering subtitle cards via chrome, and compositing with ffmpeg. supports custom tts backends and prebuilt image directories.

vibe-coding-guide codifies a workflow for structured ai-assisted development, emphasizing test-first methodology, modular decomposition, and iterative validation. covers 20 sub-skills and seven case studies spanning dashboards to subtitle generation.

auto-deploy orchestrates git-to-linux deployments with numbered build and ssh steps, backup snapshots, and health checks. covers nodejs and java projects but lacks explicit failure recovery and decision logic.

gov-procurement-advisor reviews chinese government procurement documents for regulatory compliance, scanning qualification criteria, evaluation factors, fair competition rules and contract terms against procurement law. outputs risk-flagged modification checklists with legal citations.

jx-competitor-analysis covers seo and geo competitor intelligence including keyword rankings, content strategy, backlink profiles, and ai citation patterns. outputs actionable gaps and competitive opportunities but lacks concrete data source specifications.

oo-microsoft-clarity wraps microsoft clarity api calls via oo cli with server-side credential injection. covers live insights export with optional breakdowns. read-only in current state.

validate-agent provides a structured process for checking tech choices against best practices via rag-judge precedent analysis and web research, but lacks specific trigger phrases and concrete edge case handling for validation failures.

skill-sec-scan provides static rule-based detection of security threats across multiple languages, supplemented by optional llm semantic analysis to reduce false positives. covers 57 rules across four threat categories with 100-point risk scoring.

cowork-cog bridges cellcog cloud agents to local machine execution via desktop app, enabling autonomous code, terminal, and file operations without manual approval. requires api key and desktop installation.