MCP App StoreMCP App Store
from openclaw-master-skills2,046
IM-first messaging and orchestration layer for OpenClaw: topic & P2P messaging, task and pipeline workflows, delegation, and a WebSocket-first autopoll runtime.
WTT (Want To Talk) provides an IM-first, WebSocket-native messaging and orchestration layer for OpenClaw agents. It exposes @wtt commands to manage topics, P2P messages, tasks and multi-stage pipelines, and includes an autopoll runtime that keeps a WebSocket or polling loop running to ingest events and route results into IM channels. Typical uses include cross-agent task scheduling, multi-agent pipelines for code/research work, and routing runtime events back to Telegram or other IM targets.
Use WTT when you need live, collaborative agent workflows: sharing progress/notifications into IM, delegating tasks between agents, creating pipelines that chain analysis → implementation → validation, or exposing a runtime that reconnects via WebSocket with polling fallback. It's ideal for code-focused agent tasks and multi-user shared agents.
install_autopoll.sh, status_autopoll.sh, start_wtt_autopoll.py).@wtt subcommands (config, bind, list, join, publish, poll, task, pipeline, p2p) and runtime setup (.env, agent registration, gateway tool permissions).Designed for OpenClaw agent runtime with session tools enabled; integrations reference OpenClaw gateway session tools and a Web UI binding flow. Compatible primarily with OpenClaw (agent handler/runner style) and agents that support WebSocket/polling runtimes.
WTT (Want To Talk) is an agent messaging and orchestration skill for OpenClaw with topic/P2P communication, task/pipeline operations, and WebSocket-first autopoll runtime. The SKILL.md is extremely comprehensive with detailed command references, security model docs, and setup flows. The install script is well-written but failed in audit because start_wtt_autopoll.py wasn't bundled (it's in the repo root, not scripts/). Status and uninstall scripts ran cleanly. Main security concern: the install script auto-patches OpenClaw gateway config to grant itself session tool permissions and restarts the gateway, which is invasive.
Well-structured skill with thorough documentation. The gateway config auto-patching is the most notable security concern — it modifies local OpenClaw configuration to grant the skill sessions_spawn/sessions_send/sessions_history/sessions_list permissions. While documented, this happens without explicit user confirmation during install. The skill connects to waxbyte.com (WTT's own service) which is expected for its functionality. No exfiltration, no hardcoded secrets, no destructive commands. The install script failure in audit was due to the runtime Python file not being in the scripts/ directory — a packaging issue rather than a code quality one.
Prompt Engineering Patterns
A practical collection of prompt engineering techniques and patterns for building reliable, controllable LLM prompts: CoT, few-shot, structured outputs, templat
Popup CRO
Design and optimize popups, modals, banners, and slide-ins to increase conversions without harming user experience or SEO.
Bagman — Secure Key Management
Secure key-management patterns and operational guardrails for AI agents that need to handle private keys, session keys, and secrets safely.
Atlassian Confluence Expert
Manage Confluence spaces, templates, permissions, and documentation workflows; create page hierarchies, macros, and templates for knowledge bases.
Superfluid Protocol Knowledge Base
Comprehensive developer-oriented reference for Superfluid Protocol: ABIs, forwarders, SDKs, automation, and common gotchas for building streaming-money apps.
JACS Cryptographic Provenance
Post-quantum cryptographic signing and verification for agents: sign documents, verify provenance, claim HAI.ai usernames, and send/receive signed email.
KIPRIS CLI
CLI wrapper around KIPRIS Plus OpenAPI to fetch Korean patents, trademarks, and designs and return normalized JSONL for agent consumption.
Video Call Agent
Initiate real-time video calls with a custom AI avatar (Runway Characters) that opens the conversation and returns a transcript and recording for follow-up.
listenhub
Explain anything — turn ideas into podcasts, explainer videos, or voice narration. Use when the user wants to "make a podcast", "create an explainer video", ...
InkOS - Autonomous Novel Writer
CLI agent for end-to-end fiction writing: draft, audit (33-dimension), revise, import/export and multi-genre support (English & Chinese).
Starlark Dev
Create, debug, and test Kurtosis Starlark packages — write package structure, run dry-runs, and inspect plan execution for reliable orchestration.
Maestro Orchestrate
Orchestrates complex engineering workflows with enforced design and approval gates, child-agent delegation, and session state management.
Run Execute
Orchestrates execution of work items across modes (autopilot, confirm, validate) with scripted init/complete tooling, plan/test/report artifacts and strict gati
Websocket Client Creator
Generates and guides websocket client integrations: configuration, code samples, best practices, and validation for production-ready websocket clients.
Unitree Robot Controller
Control and command Unitree robots (GO1/GO2/G1/H1) via OpenClaw: initialization, basic motion commands, and sensor integrations.
mog — Microsoft Ops Gadget
Command-line toolkit for Microsoft 365 (Mail, Calendar, OneDrive, Contacts, Tasks, Word, PowerPoint, Excel, OneNote) with slug and multi-account support.
Overnight — Autonomous Long-Running Coding
Orchestrates long-running coding goals: decomposes objectives into atomic tasks, dispatches isolated worktree workers, verifies acceptance criteria, and merges
tmux Remote Control
Control tmux sessions for interactive TTYs: send keystrokes, capture panes, orchestrate parallel coding agents, and monitor output safely.