Runs an adversarial teardown of a founder, KOL, company, or product you envy — separating marketing bubble from real signal and mapping the validated playbook onto your own resources.
Manage team goals, parked tasks and KPIs as Markdown files whose paths encode owner and status inside the Deeplake virtual filesystem.
Drive an already-logged-in Chrome through the DevTools Protocol to open pages, run JavaScript, and extract data or auth tokens.
Drives a disciplined discover-inspect-plan-query-validate loop over ktx-connected databases so analytical answers are correct, not just runnable.
Turns WeChat group chat logs pulled via wx-cli into a structured digest, with persistent history, member profiles, and group fact memory.
A release gate skill that fans out 16 parallel agents to audit every unpublished change and returns a SAFE/CAUTION/RISKY/BLOCK verdict before you npm publish.
A goal-driven execution loop that decomposes long work into atomic steps and only marks them done with real, tree-bound evidence.
Step-by-step skill for swapping a NanoClaw agent group's backend from the Anthropic SDK to OpenCode (OpenRouter, DeepSeek, Google, Zen, and more).
Installs the rtk token-compression proxy into agent containers so every Bash call is filtered for 60–90% token savings.
An interactive skill that adds channels, MCP tools, personas, or deployment changes to a NanoClaw install at the correct architectural layer.
An operations skill that wires messaging channels to agent groups in a nanoclaw instance and manages isolation levels and defaults.
Guides you through creating your first NanoClaw DM agent and verifies delivery with a welcome message.
An isolated-environment QA skill that proves omo Codex plugin hooks, installer, and TUI work without ever touching your real ~/.codex.
A durable goal-execution loop that decomposes long work into atomic steps and closes them only with real, observable evidence.
Give it a bump type and it drives the whole release: trigger the GitHub Actions publish workflow, draft release notes, announce on Discord, and verify npm.
Installs OpenAI's Codex app-server as a first-class agent backend for NanoClaw, switchable per group.
A research skill that hunts public GitHub repos and skill marketplaces for SKILL.md files, classifies each repo by archetype, and distills reusable processes and plugin ideas.
Runs an entire AI research project autonomously with a two-loop experiment/synthesis engine, from literature survey to paper draft.
A battle-tested playbook mapping real Terraform provisioner, multi-environment, and first-deploy failures to copy-paste fixes.
Runs parallel AI agents to audit your product's UX, API, architecture and docs, then synthesizes a prioritized optimization report.