ai-agent automates facebook page posting and token lifecycle management via graph api, supporting text and image content with rate-limit awareness and retry patterns for production use.

dcc-rest-gateway enables agents to control maya, blender, houdini, photoshop, and 3ds max via http rest api without mcp client. requires curl, gateway at dcc_mcp_gateway_url, and mandatory instance inventory before tool discovery or invocation.

self-improvement logs learnings, errors, and feature requests to markdown files organized by type, enabling continuous improvement through regular review and promotion of insights to broader project documentation.

product-prd-generator structures multi-chapter product requirements documents for ai-driven development, emphasizing self-contained feature modules with embedded entities, state machines, permissions, mermaid flowcharts, and business rules. each module includes field lists, flow diagrams, ui wireframes, and acceptance criteria.

oo-payhip wraps the payhip connector via oo cli, covering coupon management (create, read, update, delete), license key operations (enable, disable, verify, decrease uses), and read operations across the payhip api surface.

nm-conserve-decisive-action provides a decision framework for when to ask clarifying questions versus proceed autonomously. uses reversibility and ambiguity thresholds with a decision matrix and safety mechanisms to reduce unnecessary interruption while protecting high-stakes operations.

db-explorer-lg connects to postgresql, mysql, sqlite, mongodb, and redis to run queries, inspect schemas, and export data. includes safety guardrails, diagnostic queries, and backup workflows across all supported databases.

skill-to-agent-converter transforms openclaw skills into functional agents with proper workspace setup and session management, addressing common binding and workspace isolation failures.

blueai-models covers configuration and selection of 100+ ai models (claude, gpt, gemini, deepseek, qwen, and others) via unified bluai relay, with quick-start scripts, image generation workflows, and model selection tables for openclaw integration.

x-management implements a draft-first workflow for x/twitter account operations, emphasizing explicit user approval before any publish action. covers reads, drafts, replies, quotes, and threads with multi-account support and clear safety boundaries.

trusted-clawhub-install-gate audits local skill artifacts before installation, classifies findings as pass/review/block, assesses active-project impact, and gates installation behind explicit user approval with staged safety defaults.

whoop-health-sync exports wearable recovery, sleep, strain, and workout metrics to markdown files via oauth2 flow. covers multi-day backfills and weekly rollups with cron integration.

frontend-ui-pipeline structures vague frontend ideas into actionable briefs, design direction, build plans, and iteration roadmaps. orchestrates planning, review, and next-step clarity across the full ui workflow.

text-to-carousel converts text, articles, or urls into multi-slide social media graphics across instagram, linkedin, tiktok, and xiaohongshu using gemini image generation with explicit cjk text handling and brand consistency controls.

zvec-local-rag-service provides a local semantic search backend via http endpoints, wrapping zvec + ollama embeddings with launchd automation on macos and manual startup elsewhere.

prepares dataify builder requests for github repository scraping via url, guiding users through tool selection, parameter collection, and curl command generation with api token validation.

openmm-exchange-setup provides numbered steps to configure api credentials across mexc, gate.io, bitget, and kraken, including environment variable setup, connection verification, and basic troubleshooting for common authentication failures.

agnes-image-video wraps agnes ai's multimodal api for text-to-image, image-to-image, text-to-video, and image-to-video generation. supports compositing multiple images and includes async video polling with frame validation.

agent-analytics enables product teams to query funnels, retention, paths, and experiments via cli, with consent-based tracking setup and portfolio-aware cross-project analysis, bounded by free-tier read caps and instrumentation coverage constraints.

tokenrouter-generate-video guides agents through channel discovery, model auto-configuration, and video generation task submission across five supported models (MiniMax-Hailuo-2.3, Kling v3/v2-6, Seedance variants) with detailed parameter validation per model.

lazada-shopping-assistant automates cross-regional shopping via browser scraping, extracting live product data (prices, sold counts, shipping origins, lazmall badges) and ranking results into tiered recommendations with direct product links. supports six regional lazada sites with scenario-driven search query generation.

ec-multi-agent-setup guides creation of clawdbot agents through workspace scaffolding, identity definition, and gateway config integration. covers manual and scripted paths with validation checkpoints and rollback procedures.

imgur-cli wraps the imgur v3 api for agents to upload images from files or urls, fetch metadata, delete by hash, and manage albums. outputs json for downstream integrations like instagram, discord, or reddit posting.

news-gameinfomer fetches game informer rss feeds by section (news, reviews, previews, features) and surfaces headlines as markdown links, omitting empty sections and raw html. suitable for gaming news aggregation.

concierge-sdk is a python framework for building mcp servers and agentic applications, offering staged tool disclosure, shared state management, semantic search for large apis, and deployment to stdio or http transports.

memento provides local persistent memory for openclaw agents via sqlite-backed conversation capture, llm-powered fact extraction, and configurable recall with privacy controls. data stays on-machine by default, with optional cloud extraction explicitly opt-in.

feishu-bridge connects feishu bots to clawdbot via local websocket without requiring public infrastructure. covers bot registration, credential setup, bridge startup, macos service auto-restart, and group chat trigger tuning.

china-technical-analysis computes macd, kdj, rsi, moving averages, and bollinger bands for chinese futures markets, generating buy/sell signals weighted by indicator agreement and support/resistance levels.

supply-chain-intel aggregates freight rates, port data, trade flows, and risk signals across 14 factors and 4 modes, delivering cost optimization and disruption alerts via structured workflows and multi-source validation.

micro-influencer-finder provides a systematic 7-step framework for discovering and vetting creators in the 1k-100k follower range, covering audience authenticity checks, cost-per-engagement modeling, and roi tracking. includes detailed examples and common failure modes.

bit provides a cli entry point for the bit url shortener, enabling creation, querying, updating, deletion, and click analytics on short links through straightforward command invocation.

nervix-onboarding covers end-to-end agent enrollment and skill bundle publishing within the nervix federation, including prerequisite validation, cli-based and app-based enrollment flows, and clawhub readiness checks before publishing.

estimate-builder-qmohd constructs detailed construction cost estimates with configurable line items, category-based rollups, and markup calculations. exports to excel and validates inputs before finalization.

axodus-frontend-builder scaffolds react, next.js, or vite applications with api client patterns, accessibility guardrails, and quality validation. targets repeatable frontend builds with safe credential handling and error resilience.

yt fetches youtube transcripts, searches videos and channels, and lists recent uploads via transcriptapi. covers transcripts, search, channel metadata. no upload or account mgmt.

safe-config-workflow-zh-cn guides openclaw configuration changes through pre-confirmation, automated repair, backup comparison, and gateway validation, with explicit filtering rules for operator feedback.

baml-codegen generates type-safe llm extraction code from natural language via baml, supporting multimodal inputs and 10+ frameworks with mcp-backed pattern matching and 95%+ compilation success.

obsidian-curator manages obsidian vaults through couchdb/livesync, offering capture, ai-enriched processing, filing, task management, and vault auditing via cli or node.js api. works offline with optional cloud ai providers.

free-feishu-voice generates tts audio via espeak or edge-tts, uploads to feishu, and sends voice messages to configured chats. credentials and voice options are externalized to config files and environment variables.

trmnl generates and sends html content to e-ink displays via cli, supporting flexbox, grid, and typography utilities within webhook payload limits. validates before send and maintains send history.

sifs-search locates code by intent, symbol, or behavior across local checkouts and git sources. designed for codebase exploration before broad file reads, call-site discovery, and architecture tracing in agent runtimes with shell access.

nvidia-cuda covers gpu-aware pytorch training and inference on modern tensor core systems, with concrete checks for device placement, dtype policy, distributed execution, and profiling before kernel work.

openclaw-enhanced-memory replaces monolithic memory with hierarchical directories, tag-based indexing, automatic lifecycle archival, and smart cross-category retrieval. three python scripts handle search, routing, and age-based organization.

tencent-tts-podcast converts text to podcast audio via tencent cloud tts, handling both short and long text with automatic chunking, parallel processing, and optional cos upload. supports 26 chinese voices and generates up to 30-minute audio files.

codex-reset-radar polls a monitoring endpoint hourly to detect codex quota reset windows, comparing json state and pushing chat alerts only when changes occur. uses stdlib python and integrates with openclaw cron for minimal token overhead.

hy-mt2-translator-skill wraps the hunyuan translation model with support for 38 languages across 6 modes including terminology constraints, style control, and structured data preservation. supports both tencent cloud and self-hosted backends with credential persistence.

store-policy-writer generates tailored ecommerce policies (returns, shipping, privacy, terms) grounded in actual business operations rather than generic templates, with explicit dispute-prevention detail on refund timing, return shipping costs, and exclusions.