Loading...
Loading...
Found 90 Skills
Turn the current conversation context into a specification document. Use when user wants to create a lean specification from the current context.
Orquestra o Herdr e os agentes de código nas duas superfícies, Herdr TUI (comando herdr) e Herdr Desktop com herdr-core (comandos hc e herdr-app). Use quando a pessoa pedir para configurar o Herdr, localizar o agente nas janelas, abrir painéis, abas, espaços ou workspaces, iniciar agentes Claude Code, Codex ou OpenCode com as permissões liberadas, enviar tarefas, ler a tela de outro agente, esperar resultados, rodar comandos demorados ou em outra máquina pelo herdr-core, limpar terminais ou gravar as instruções do Herdr no AGENTS.md e no CLAUDE.md.
Measure and improve how coding agents discover and use your product with Gauge Agents. Use when asked to run agent evaluations, test whether agents can use an SDK, API, CLI, MCP server, docs, or skill, find where agents struggle with onboarding, measure which tools agents choose, set up Gauge PR checks for a repository, or read and explain Gauge Agents results, sessions, and failed Gauge checks. Works through the Gauge CLI or the Gauge MCP. Not for ordinary unit tests or AI-search visibility tracking.
Writes code to deliver a specified feature or task in an AFK loop
Multi-model coding harness on OpenCode subagents. The orchestrator designs and specs, models picked by routing build, another maker reviews.
Coordinate multiple coding agents through Herdr panes in one workspace. Use when the user asks to plan, delegate, supervise, review, or dogfood work with OMP or other agents through Herdr. Requires HERDR_ENV=1.