How do I install the Agentdoctor CLI skill?

Run npx skillmds@latest add sebastianboehler/agentdoctor-cli in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agentdoctor CLI safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agentdoctor CLI?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agentdoctor CLI free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agentdoctor CLI?

SebastianBoehler (@sebastianboehler) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agentdoctor CLI by sebastianboehler

Name: Agentdoctor CLI
Author: SebastianBoehler

Agentdoctor CLI

Check host dependencies with the local `agentdoctor` CLI. Use when an agent workflow fails because a required tool such as an SMB client, SSH client, or HTTP fetch utility is missing or when a task needs a machine-readable preflight report.

SebastianBoehler f616650 2 files · 1.8 KB Updated Sep 17, 2026

File contents

AgentDoctor CLI

Use agentdoctor to diagnose missing host capabilities before a workflow starts or immediately after a dependency-related failure.

Invoke the Tool

Prefer the installed binary when it exists:

agentdoctor -profile smb-client -format text

If working inside this repository or the binary is not installed, run:

go run ./cmd/agentdoctor -- -profile smb-client -format text

Choose Profiles

Use -profile smb-client when a workflow needs SMB or CIFS access to a Windows host or share.
Use -profile ssh-client when the task depends on SSH or SCP.
Use -profile web-fetch when the task depends on curl or PowerShell HTTP support.
Use repeated -cmd flags for custom one-off dependency checks.

Apply Good Defaults

Keep -strict=true for startup and health checks so missing tools fail early.
Use -format json when another agent step will branch on the result.
Use -strict=false when the task only needs advisory diagnostics.
Prefer agentdoctor over a blind failing connection attempt when the problem may be “tool missing” rather than “credentials wrong.”

Examples

Diagnose the SMB client case:

go run ./cmd/agentdoctor -- -profile smb-client -format text

Check a custom command set:

go run ./cmd/agentdoctor -- -cmd git -cmd curl -format json

Frequently asked questions