Pular para o conteúdo
TakoAPI
Voltar às skills

autonomous-execution

por pouyakhodadust-eng

Productivity & Tasks
835Downloads informados pelo clawskills.sh
0Visualizações
0 curtidaEntre para curtir esta skill
—Ainda sem avaliações
Sua avaliação

Resumo

Execute tasks end-to-end while respecting safety boundaries.

Descrição

O que esta skill faz

Defines how an AI agent should handle multi-step tasks: complete subtasks sequentially without prompting, but stop and ask before sending messages, deleting data, accessing credentials, or making payments. Error handling distinguishes between safe retries and situations that need user input.

Instead of agents that halt at every step or run unchecked, this provides a clear behavioral contract: finish routine work automatically and gate only on operations with irreversible side effects.

Quando usá-la

  • Batch-processing a list of research topics without repeated check-ins
  • Running sequential file analysis tasks and getting a single completion report
  • Fetching data from multiple read-only APIs and compiling results
  • Automating a workflow that ends with a confirmation gate before external output
  • Retrying rate-limited API calls without manual intervention

Exemplo de fluxo de trabalho

ENTRADA

User asks: Research these 5 API documentation pages and email a summary to team@company.com

AGENTE
  1. 1

    List all subtasks: read 5 documentation pages, compile summary, send email

  2. 2

    Fetch and read each of the 5 documentation pages sequentially

  3. 3

    Compile a combined summary from all sources

  4. 4

    Stop before sending email and request user confirmation

SAÍDA

Research complete. Confirm: send summary to team@company.com?

Instalação

clawhub install pouyakhodadust-eng/autonomous-execution

Execute o comando da CLI ou cole o prompt no Claude Code, Cursor, Codex ou no agente de código que você usa.

Links