social-ops decomposes social media campaign execution into six discrete roles (scout, researcher, content specialist, responder, poster, analyst) with documented handoff points and a configurable data directory contract. designed for auditability and composability across platforms.

polymarket-edge-trader scans active aion polymarket markets, identifies highest-edge opportunities using a user-supplied fair probability, applies kelly-style position sizing, and executes trades with built-in safeguards for risk alerts, slippage, and flip-flop detection.

oo-databox wraps the databox connector via oo cli, exposing 7 actions for crud operations on data sources and datasets. requires prior oomol account setup and databox api key connection.

akshare-cn-market wraps chinese stock market and macro data (gdp, cpi, pmi, m2, bond yields) via akshare library. provides cli scripts for hist klines, indices, financial summaries, and sentiment data (limit-up pools, dragon-tiger boards, fund flows). acknowledges network restrictions and data delays.

mineru-pdf extracts text, tables, and formulas from pdfs via mcp with optional apple silicon acceleration. two invocation paths exist: direct tool for file persistence, mcp method for claude integration. output is markdown with embedded latex formulas.

fullrun-cli wraps google ads campaign management through a json-returning command interface. supports triage-driven diagnosis and ai-automated optimization with explicit rate limits and guardrail enforcement.

moltguild enables agents to register, claim bounties, deliver work, and earn usdc on solana with x402 escrow. includes reputation tracking, raids, wallet setup, and webhook notifications. registration yields free sol airdrop.

podcast-to-video-maker converts audio files into captioned 1080p mp4 video via remote gpu rendering. users upload mp3/wav/m4a files and describe desired output (captions, waveforms, overlays) to get shareable video within 1-3 minutes without manual editing.

background-music adds ai-generated background tracks to video files (mp4, mov, avi, webm up to 500mb) and exports 1080p results in 30-90 seconds via cloud gpu rendering. requires nemo_token authentication and session management.

swarms-ai covers single and multi-agent orchestration via the swarms api, supporting 20+ architecture patterns from sequential workflows to hierarchical swarms. includes token launch on solana, marketplace integration, and streaming responses across python, typescript, and curl.

robotaxi-briefing tracks autonomous driving sector intelligence across pony.ai, waymo, tesla, and baidu apollo via twice-daily incremental reports. combines rss feeds, web search, reddit, and direct page fetching with 24-hour delta extraction and url resolution.

guides step-by-step import of existing alibaba cloud resources into terraform management across nine phases, supporting both one-time migration and incremental sync with resource discovery, hcl generation, and state import automation.

moltawards-revenue-hunting-for-ai-agents exposes federal contracts, state bids, grants, and jobs via REST API with naics-scoped feed ranking and adjacency narratives. text-only platform for agent-driven opportunity triage and pursuit teaming.

amazon-launch-playbook generates a 90-day product launch strategy covering category evaluation, listing optimization, inventory management, and phased advertising with weekly milestones and success metrics.

pdf-ocr-tool converts pdfs and images to markdown using ollama glm-ocr with content-type detection for text, tables, and figures. supports mixed mode region splitting and customizable prompts.

react-component-auditor covers performance bottlenecks, hook bugs, accessibility gaps, and prop design issues in react codebases. detects memo opportunities, stale closures, and state management anti-patterns with categorized findings.

text skill covers common transformations like encoding normalization, whitespace cleanup, and regex extraction with reference commands. lacks domain-specific procedures for the five advertised categories (writing, data, academic, copy, localization).

text-to-video-animation-ai converts text prompts into 1080p mp4 animations via cloud gpu processing. supports txt, docx, pdf, png inputs up to 200mb with 1-2 minute turnaround. requires nemo_token authentication and session management.

blog-to-kindle fetches essay archives from supported sites (paul graham, kevin kelly, wait but why, et al), generates ai cover art, compiles into kindle-formatted epub, and sends via mail. supports incremental updates via state tracking.

maker-free-google wraps a cloud video rendering backend, accepting images or clips via upload and generating finished 1080p mp4s from natural language requests without ui interaction. targets content creators and small businesses seeking fast, free video composition.

ccapi wraps 60+ ai models (gpt, claude, gemini, deepseek, sora, kling) behind openai-compatible endpoints. useful for multi-provider text, image, video, and audio generation without sdk switching.

yijian-ba-baokuan extracts video text from douyin, bilibili, and xiaohongshu links, analyzes why they perform, then outputs three platform-specific rewritten versions ready to publish. targets content creators doing rapid multi-platform distribution.

backlink-analyzer audits link profiles for quality, toxicity, and competitive gaps. it maps to domain-authority-auditor for formal risk scoring and surfaces opportunities via competitive intersection and anchor text analysis.

unformal-notifications covers three notification patterns (scheduled routine, local listener, on-demand api call) for tracking pulse responses. desktop routine option runs locally with timestamp-based deduplication, listener provides real-time macos notifications, on-demand option requires no setup.

obsidian-charts-layout covers charts view plugin setup and g2plot syntax for pie, bar, column, line charts in obsidian with annotation labels, including troubleshooting and python file-writing patterns for vault integration.

chart-animation renders time-series data as dark-themed gif/mp4 animations and static overviews, supporting up to 4 series with configurable frame rates and dimensions.

codebuddy-cli covers installation, configuration, and command usage for tencent's ai-powered terminal coding assistant. supports natural language prompts, permission controls, and region-specific model selection via login method.

automated daily news briefing system that aggregates tech and finance articles from baidu or duckduckgo, applies ai commentary, and delivers via qq, telegram, or discord on a configurable schedule.

absolute-tao maps osho's commentary on lao tzu to five modern scenarios (anxiety, effortless work, letting go, inner stillness, authenticity). trigger phrases are specific and routing is explicit, but step-by-step execution remains vague and failure modes unaddressed.

hutian-opc-pre-publish-check guides skill/agent authors through nine standardized quality gates including case coverage, consistency testing, competitive positioning, and compliance verification. comprehensive but procedurally dense.

alab-skills is a routing guide for the ALab experiment workbench, explaining the three-layer role hierarchy (admin/project/worker), credential hygiene rules, and when to delegate to narrower subskills rather than a procedural skill itself.

feishu-notes provides oauth-backed document creation, reading, and formatting in feishu cloud drive via api. setup is manual and multi-step. lacks explicit error handling and recovery guidance.

a-brief-history-of-intelligence maps max bennett's neuroscience framework onto ai development, routing five intent categories (evolutionary breakthroughs, first brains, mammalian cognition, human language, ai parallels) through referenced materials with consistent watermarking and language preservation.

openclaw-ghost-pay wraps x402 payment discovery, execution, and settlement reporting for Ghost protocol merchants, plus optional GhostWire escrow flows. requires signer key and nodeenv setup.

phy-reddit-cultivate automates reddit account growth via applescript-controlled chrome, scanning rising posts and drafting comments without detection. targets indie developers and requires macos with manual browser setup.

generates diagrams in mermaid, plantuml, or ascii from natural language descriptions. covers flowcharts, sequences, architecture, er models, and state diagrams with formatting guidance and rendering options.

vidix-ai-photo-video-generator converts jpg photos into 1080p mp4 videos via cloud gpu rendering. users upload images, describe desired output, and download within 30-90 seconds. supports transitions, music, text overlays, and multi-track editing through a session-based api.

ai-video-creator-online converts images and text into 1080p mp4 files via cloud gpu rendering in 1-2 minutes. supports mp4, mov, jpg, png up to 500mb. requires nemo_token authentication and handles session management, uploads, and exports through a rest api.

podcast-pipeline covers interview prep, post-recording show notes with chapter generation and clip suggestions, and solo episode planning with platform-specific social outputs. includes episode history and sponsor tracking via persistent json storage.

auto-body-collision-repair-kit enforces seven regulatory gates (epa neshap, i-car certification, parts disclosure, insurance drp, warranty accuracy, betterment, ftc review claims) to block common marketing violations in collision repair. includes four prompt files covering seasonal campaigns, service pages, reputation management, and digital ads.

foreseekai converts natural language predictions into kalshi market trades, supporting parse, execute, and position management via api. covers elections, economics, crypto, sports, and weather outcomes with rate-limited token budgets across free, pro, and ultra tiers.

comic-create converts user ideas into executable manga planning documents: scripts, storyboards, character designs, dialogue, and layout specs ready for artists and downstream pipelines. it prompts for missing context before proceeding.

kingdee-data-exporter automates bulk export of kingdee k3cloud operational data (bills and reports) to excel via command-line interface with organizational filtering and secondary filtering utilities.

local-whisper-hardened wraps openai whisper for offline speech-to-text transcription with configurable model sizes from 39m to 1.5gb. includes command-line interface with json output and timestamp support.

text-to-video-notegpt converts uploaded notes (txt, docx, pdf, md) into 1080p mp4 videos via cloud rendering. users describe desired output and receive finished video without manual editing or recording.

game-design-attribution-audit provides a framework for evaluating how players interpret success and failure through locus, stability, and controllability. structured around player perception rather than mechanical correctness, it diagnoses fairness and churn risk by reconstructing player experience and mapping attribution profiles to design interventions.

the-power-of-now maps eckhart tolle's framework to five common mental patterns: mind identification, pain-body dissolution, presence practice, ego in relationships, and surrender. it routes user queries through intent tables and references, with structured trigger phrases and watermarking rules.