Analyzes your article, decides where illustrations belong, and generates a visually consistent set of images via a Type × Style × Palette system.
Designs and generates article cover images using a 5-dimension system of type, palette, rendering style, text level, and mood.
Publishes Markdown, HTML, or plain-text drafts straight into your WeChat Official Account (微信公众号).
A diagnose-first Terraform/OpenTofu companion that classifies the failure mode, applies version-aware guards, and always ships validation commands plus rollback notes.
Audits a paid ad account against a 20-item ROAS rubric and returns a SHIP/FIX/BLOCK verdict before you launch or scale spend.
Run several Claude Code sessions as a lead-plus-teammates crew with a shared task list, mailbox messaging, and file locks.
Wires a personal WeChat account into NanoClaw using Tencent's official iLink Bot API with QR-code login.
A step-by-step, approval-gated runbook for moving legacy NanoClaw and Claude-native memory files into the shared `memory/` tree.
Spins up 5 mutually hostile agents for a 3-round critique debate, keeps only the defensible insights, then hands them to the `plan` agent for an executable plan.
Maps your codebase into per-feature Mermaid flowcharts, proves where logic is duplicated, and proposes the simplest unified architecture.
A skill that detects your project's languages, installs and configures the right language server, then verifies it with a real diagnostics roundtrip.
A three-tier escalation skill that reaches pages blocked by 403, Cloudflare, JS-only rendering, or login walls.
Type 'ultrawork' in your prompt and the agent switches to a maximum-precision mode that demands failing-first proofs, real-surface evidence, and cleanup receipts before claiming done.
Scans and scores your repository, then auto-generates AGENTS.md files at the root and in the subdirectories that actually deserve one.
Installs NanoClaw's native Baileys WhatsApp adapter and links your device via QR code or pairing code.
Analyzes your whole repo and generates a hierarchical AGENTS.md knowledge base at the root plus high-complexity subdirectories.
A QA-focused skill for smoke-testing opencode's CLI, HTTP server, plugin hooks/events and TUI, plus read-only session DB investigation.
Typing 'ultrawork' or 'ulw' flips the agent into a maximum-precision mode that demands failing-first proofs, real-surface QA evidence, and cleanup receipts before it can claim done.
Spins up five mutually hostile agents for three rounds of cross-critique, then turns only the surviving insights into an executable plan.
Spawns five mutually hostile agents to cross-critique your plan for three rounds, then hands only the surviving insights to a dedicated planner.