OpenParable
Skip to content
← Back to skills

Netcatty Tool CLI

Apps & Automation
netcatty-tool-cli

Netcatty Tool CLI is an Agent Skill for apps & automation published in the binaricat/Netcatty repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Use this skill when an external agent needs to operate on Netcatty sessions through Skills + CLI instead of the netcatty-remote-hosts MCP server.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/binaricat/Netcatty/blob/HEAD/skills/netcatty-tool-cli/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/netcatty-tool-cli, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Netcatty Tool CLI

Use this skill for external ACP agents when Netcatty is configured for Skills + CLI mode.

For routine tasks, the host prompt is usually enough. Read only the reference that matches the task type.

Router

  1. Use the exact Netcatty CLI prefix provided by the host prompt.
  2. Keep --chat-session <chat-session-id> on every Netcatty CLI call. Do not omit it.
  3. Treat --chat-session <chat-session-id> as required for env, session, real exec, and every sftp operation. Treat --session <session-id> as required for session, exec, and every sftp operation.
  4. Classify the task before choosing a command path:
    • Remote command execution tasks go through the exec reference.
    • Remote file or directory tasks go through the sftp reference.
    • If the user explicitly says to avoid shell or exec, do not use exec.
    • Treat exec as the short-command path only. If the command may exceed about 60 seconds, or streams output for an extended period, use the long-running job commands instead of plain exec.
  5. If the host prompt already names a connected default target session, use that session directly for routine requests that do not mention another session or host, but still start with session --session <id> --json --chat-session <chat-session-id> instead of jumping straight to exec or sftp.
  6. Only fall back to env lookup when the task is ambiguous, the user points to another session, or that direct session lookup fails.

Core Rules

  • Treat the host-provided CLI prefix as the only supported entrypoint for this session.
  • If a command launcher is needed, prefer the operating system's built-in launcher for the current environment; do not require optional shells that may not be installed.
  • Run Netcatty CLI commands strictly serially.
  • Treat Netcatty CLI errors as authoritative.
  • Never ask the user for SSH credentials, key paths, proxy settings, or jump-host details when Netcatty session access already exists.
  • Do not pause to explain the plan, re-read this skill, or design scripts before trying that shortest path.
  • When presenting structured results, prefer a concise table if it fits clearly.

Examples:

  • On Windows, if a literal shell command line is required, use the host-provided prefix with the system launcher available in the environment, such as cmd.exe or Windows PowerShell; do not assume PowerShell 7 pwsh.exe exists.
  • On macOS or Linux, use the host-provided prefix directly, or the system shell already available in that environment when a shell command line is unavoidable.
  • When the execution surface accepts argv-style calls, use the Netcatty launcher path as the executable and pass subcommands and flags as separate arguments instead of wrapping it in another shell.

References

  • Exec and session workflow: references/exec.md
  • SFTP file workflow: references/sftp.md
  • Session and device-type handling: references/session-types.md
  • Cancel, resume, and runtime diagnostics: references/control-commands.md
  • Error handling and authoritative failures: references/errors.md