RouterOS recovery and reinstall workflows are easy to misorder. This skill gives grounded Netinstall guidance for packages, scripts, interfaces, and first boot behavior.

RouterOS discovery is hard to implement when MNDP packet behavior is unclear. This skill explains the wire format, neighbor commands, parsing details, and safety constraints.

RouterOS hotspot setup has hidden rules around firewall chains, CAPPORT, RADIUS, and template variables. This skill guides Claude, Codex, and Claude Code through accurate captive portal configuration.

RouterOS differs from standard Linux, so generic shell advice often fails. This skill gives Claude, Codex, and Claude Code RouterOS-specific CLI, REST API, and automation guidance.

RouterOS firewall syntax is easy to misorder and duplicate. This skill gives focused guidance for filters, NAT, mangle routing, lists, and DoS rules.

RouterOS containers require device-mode, storage, VETH networking, and version-specific commands. This skill helps Claude, Codex, and Claude Code produce safer MikroTik container workflows.

RouterOS command trees are hard to inspect and convert into schemas. This skill explains /console/inspect traversal, REST verb mapping, and safe schema generation.

RouterOS /app YAML differs from docker-compose and changes across RouterOS releases. This skill explains the format, validation choices, REST loading patterns, and common mistakes.

Technical documents often lose precision when prose, commands, paths, and structured data are edited inconsistently. STOW applies output contracts and focused prose review while preserving protected literals.

Managing Tencent Meeting from an AI assistant requires many commands, identifiers, permissions, and confirmation steps. This skill routes requests through tmeet with documented privacy and safety checks.

Frontend work often fails when framework conventions, browser validation, and deployment rules are mixed together. This skill gives agents a structured path for React, Vue, Vite, and CloudBase web work.

Generic AI interfaces often lack visual direction and platform-aware detail. This skill guides Claude, Codex, and Claude Code to define design intent before building prototypes.

Large coding changes often fail when scope, design, and acceptance criteria are vague. This skill guides Claude, Codex, and Claude Code through requirements, design, tasks, and staged confirmation.

Browser apps need a clear CloudBase relational database setup path. This skill shows the shared web SDK client and Supabase-style query flow.

CloudBase MySQL operations need clear tool choice and safety checks. This skill guides agents through MCP-based provisioning, querying, schema changes, and permission reviews.

CloudBase PG apps need correct SDK, schema, auth, RLS, and storage setup. This skill guides agents through safe PostgreSQL workflows and verification.

CloudBase teams need a fast way to understand resource health, logs, and service status during incidents. This skill guides Claude, Codex, and Claude Code through read-only inspections and prioritized recommendations.

WeChat Mini Program work requires platform-specific structure, preview, and CloudBase decisions. This skill guides setup, debugging, release checks, and CloudBase integration.

CloudBase teams need reliable raw HTTP integrations when SDKs are unavailable. This skill guides endpoint selection, authentication, OpenAPI lookup, and safe request construction.

Complex schema work needs clear entity boundaries and relationships. This skill guides Claude, Codex, and Claude Code through Mermaid models and CloudBase data-model tool use.