How do I install the Doctor skill?

Run npx skillmds@latest add datasift-ty-personal/doctor 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 Doctor safe to use?

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

Which AI agents work with Doctor?

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 Doctor free to use?

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

Who published Doctor?

DataSift-Ty-Personal (@datasift-ty-personal) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Doctor by datasift-ty-personal — Agent Skill

Name: Doctor
Rating: 4.2 (5 reviews)
Author: DataSift-Ty-Personal

Doctor

SiftStack readiness check. Reports which installed SiftStack skills work right now, which need a login, which need an API key, what each key costs, and the no-API route to use instead. Use when someone asks "what can I run", "what do I need to set up", "why did that skill skip a step", "which keys do I need", "is SiftStack set up", "siftstack doctor", or right after installing any SiftStack plugin or bundle. Read-only, sends no requests, spends nothing, never prints a credential.

File contents

SiftStack doctor

Answer one question: of the SiftStack skills, what runs today and what is blocked on what.

Run it

python "${CLAUDE_PLUGIN_ROOT}/scripts/doctor.py"

Use python3 if python is not on the PATH. Add --env-dir <folder> when the person keeps their .env somewhere other than the current folder, and --json when you need to reason over the result instead of showing it.

The script reads environment variables and a .env in the chosen folder. It records which NAMES are set and never reads a value into its output. You MUST NOT open or print the .env yourself, because the whole point of the script is that a credential never lands in the conversation.

Report it

Give the person the three groups in this order, in plain words:

Works right now. These need nothing.
Signed-in services. Keys are fine, but the skill drives a site they must be logged in to. This cannot be checked from here, so say so instead of calling it ready.
Needs a credential. For each: the variable name, the real cost, and the route to use without it. A blocked skill with a fallback is not a dead end, so always say the fallback in the same breath.

Then offer the setup skill if anything is blocked and they want to unblock it. Do not push keys on someone whose work is covered by the first group.

When it cannot run

No Python available (some Cowork sessions): read ${CLAUDE_PLUGIN_ROOT}/data/requires.json directly. Each package lists requires.tier (none, account, api), requires.env, requires.cost and requires.fallback. Report tier none as working, and everything else as "needs X", since without the script you cannot see which variables are set.

---name: doctordescription: SiftStack readiness check. Reports which installed SiftStack skills work right now, which need a login, which need an API key, what each key costs, and the no-API route to use instead. Use when someone asks "what can I run", "what do I need to set up", "why did that skill skip a step", "which keys do I need", "is SiftStack set up", "siftstack doctor", or right after installing any SiftStack plugin or bundle. Read-only, sends no requests, spends nothing, never prints a credential.---# SiftStack doctorAnswer one question: of the SiftStack skills, what runs today and what is blocked on what.## Run it```bashpython "${CLAUDE_PLUGIN_ROOT}/scripts/doctor.py"```Use `python3` if `python` is not on the PATH. Add `--env-dir <folder>` when the person keeps their `.env` somewhere other than the current folder, and `--json` when you need to reason over the result instead of showing it.The script reads environment variables and a `.env` in the chosen folder. It records which NAMES are set and never reads a value into its output. You MUST NOT open or print the `.env` yourself, because the whole point of the script is that a credential never lands in the conversation.## Report itGive the person the three groups in this order, in plain words:1. **Works right now.** These need nothing.2. **Signed-in services.** Keys are fine, but the skill drives a site they must be logged in to. This cannot be checked from here, so say so instead of calling it ready.3. **Needs a credential.** For each: the variable name, the real cost, and the route to use without it. A blocked skill with a fallback is not a dead end, so always say the fallback in the same breath.Then offer the `setup` skill if anything is blocked and they want to unblock it. Do not push keys on someone whose work is covered by the first group.## When it cannot runNo Python available (some Cowork sessions): read `${CLAUDE_PLUGIN_ROOT}/data/requires.json` directly. Each package lists `requires.tier` (`none`, `account`, `api`), `requires.env`, `requires.cost` and `requires.fallback`. Report tier `none` as working, and everything else as "needs X", since without the script you cannot see which variables are set.

Frequently asked questions