How do I install the Doctor skill?

Run npx skillmds@latest add horizonbrute/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. 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?

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

Doctor by horizonbrute — Agent Skill

Name: Doctor
Author: HorizonBrute

Doctor

Run the Horizon AIOS health check — verifies env vars, the skills symlink, git hooks, local config, the AIOS registry, and privileged-dir Deny ACLs, then reports passed/warnings/failures. Use when the user types /doctor, asks to "run the doctor", "health check the AIOS", "check the install", or diagnoses why the AIOS is misbehaving.

HorizonBrute 4c1da6e 2.6 KB Updated Sep 17, 2026

File contents

Skill: /doctor

Model preference: #lowcost (per horizon_aios_model_prefs.md; overridable by a prompt directive).

Run the read-only Horizon AIOS health check and surface its results. Use this after bootstrap, after an upstream sync that touches bootstrapping, or whenever something behaves unexpectedly and you need a quick status read.

When to invoke

The user types /doctor, or asks you to "run the doctor", "health-check the AIOS", "check the install", or diagnose a broken/misbehaving AIOS.

Step-by-step execution

Step 1 — Run the check

python "$HORIZON_SYSTEM/sbin/horizon_aios_doctor.py"

Read-only — no elevation required. It takes no arguments and runs every check unconditionally.

Step 2 — Report the result

2.1 Relay the summary line (N checks passed, M warnings, K failures).

2.2 For each [FAIL] and [WARN], surface the line and its remediation hint (the script prints the fix inline, e.g. "run bootstrap", "run horizon_aios_harden.py", "run horizon_aios_switch.py init"). Do not invent fixes — quote the script's own guidance.

2.3 Exit code is 1 if any check failed, 0 if only warnings or clean. Report success/failure accordingly.

Post-setup mode (`--post-setup`)

Right after an install, bootstrap, or /harden, run the post-setup mode to confirm the user-facing plumbing actually works:

python "$HORIZON_SYSTEM/sbin/horizon_aios_doctor.py" --post-setup

This runs all the standard checks plus three post-install verifications: a test sound through the canonical sound chokepoint, statusline command resolution, and git commit.gpgsign. A muted sound config reports a clean [SKIP] (not a failure), and the summary then includes a skipped count. Exit-code semantics are unchanged — exit 1 only if a check FAILs; SKIP and WARN do not fail the run.

Notes for the executing agent

$HORIZON_SYSTEM must be set. If it is not, report that the AIOS environment is not active and the user should source their profile or run the AIOS switcher.
This skill is purely diagnostic — it changes nothing. A stopped monitor is a WARN, not a FAIL (the monitor is optional).
Do not reimplement any check — the script is the single source of truth. Run it and format its output.

---name: doctordescription: Run the Horizon AIOS health check — verifies env vars, the skills symlink, git hooks, local config, the AIOS registry, and privileged-dir Deny ACLs, then reports passed/warnings/failures. Use when the user types /doctor, asks to "run the doctor", "health check the AIOS", "check the install", or diagnoses why the AIOS is misbehaving.---# Skill: /doctor**Model preference:** `#lowcost` (per `horizon_aios_model_prefs.md`; overridable by a prompt directive).Run the read-only Horizon AIOS health check and surface its results. Use this after bootstrap, after an upstream sync that touches bootstrapping, or whenever something behaves unexpectedly and you need a quick status read.---## When to invokeThe user types `/doctor`, or asks you to "run the doctor", "health-check the AIOS", "check the install", or diagnose a broken/misbehaving AIOS.---## Step-by-step execution### Step 1 — Run the check```python "$HORIZON_SYSTEM/sbin/horizon_aios_doctor.py"```Read-only — no elevation required. It takes no arguments and runs every check unconditionally.### Step 2 — Report the result2.1 Relay the summary line (`N checks passed, M warnings, K failures`).2.2 For each `[FAIL]` and `[WARN]`, surface the line and its remediation hint (the script prints the fix inline, e.g. "run bootstrap", "run horizon_aios_harden.py", "run horizon_aios_switch.py init"). Do not invent fixes — quote the script's own guidance.2.3 Exit code is `1` if any check failed, `0` if only warnings or clean. Report success/failure accordingly.---## Post-setup mode (`--post-setup`)Right after an install, bootstrap, or `/harden`, run the post-setup mode toconfirm the user-facing plumbing actually works:```python "$HORIZON_SYSTEM/sbin/horizon_aios_doctor.py" --post-setup```This runs all the standard checks plus three post-install verifications: a testsound through the canonical sound chokepoint, statusline command resolution, andgit `commit.gpgsign`. A muted sound config reports a clean `[SKIP]` (not afailure), and the summary then includes a skipped count. Exit-code semantics areunchanged — exit 1 only if a check FAILs; SKIP and WARN do not fail the run.---## Notes for the executing agent- `$HORIZON_SYSTEM` must be set. If it is not, report that the AIOS environment is not active and the user should source their profile or run the AIOS switcher.- This skill is purely diagnostic — it changes nothing. A stopped monitor is a WARN, not a FAIL (the monitor is optional).- Do not reimplement any check — the script is the single source of truth. Run it and format its output.

Frequently asked questions