How do I install the Os Environment Probe skill?

Run npx skillmds@latest add richfrem/os-environment-probe 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 Os Environment Probe safe to use?

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

Which AI agents work with Os Environment Probe?

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 Os Environment Probe free to use?

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

Who published Os Environment Probe?

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

Os Environment Probe by richfrem — Agent Skill

Name: Os Environment Probe
Rating: 4.5 (4 reviews)
Author: richfrem

Os Environment Probe

Discovers and persists the user's available AI environments (Claude, Copilot CLI, Agy CLI, Cursor, etc.) to context/memory/environment.md. Run once after OS setup or whenever the environment changes. os-architect and os-evolution-planner read this file to select the right delegation backend and cheapest brainstorm model automatically. Invoked by os-architect on first run if environment.md is absent.

richfrem 0a6eb1b 9 files · 4.0 KB Updated Sep 10, 2026

File contents

Role

os-environment-probe asks the user which AI environments they have access to, then verifies each claimed environment by running a lightweight probe command. Results are written to context/memory/environment.md — a single source of truth that downstream skills read to make delegation decisions without asking the user again.

Environment Interview

Ask the user these questions (one prompt, multiple-choice, keep it brief):

Which of these AI tools do you currently have active on this machine? (Select all that apply)

A. Claude Code only B. Claude Code + GitHub Copilot CLI (Pro or Business plan) C. Claude Code + Agy CLI (Antigravity — gemini-3.5-flash backend) D. Cursor (Claude or GPT backend) E. Other (describe)

Wait for their answer before probing.

Probe Commands

For each claimed environment, verify it is actually callable:

Environment	Probe command	Pass condition
Copilot CLI	`gh copilot explain "test" 2>&1 \| head -3`	No "not authenticated" or "command not found"
Agy CLI	`agy --version 2>&1 \| head -1`	Outputs a version string
Cursor	`cursor --version 2>&1 \| head -1`	Outputs a version string
Claude Code	always present	—

Report each probe result to the user:

Pass: "✓ Copilot CLI — confirmed"
Fail: "✗ Copilot CLI — not found or not authenticated (skipping)"

Only write environments that pass to the profile.

Procedure

Run the Environment Interview.
Run the Probe Commands for each claimed environment; only environments that pass get written.
Write context/memory/environment.md in the format in references/detailed-reference.md.
Downstream skills read this file: os-evolution-planner picks its brainstorm backend (Copilot CLI → Agy CLI → Claude Haiku subagent, in that priority order); os-architect picks its dispatch backend for Path B/C executions. If environment.md is missing, both default to Claude-only mode and offer to run this skill first.
Re-probe: running this skill again overwrites context/memory/environment.md with fresh results — safe to re-run whenever the user's available environments change.

Output format, smoke tests, gotchas, and support-file details are in references/detailed-reference.md.

---name: os-environment-probedescription: Discovers and persists the user's available AI environments (Claude, Copilot CLI, Agy CLI, Cursor, etc.) to context/memory/environment.md. Run once after OS setup or whenever the environment changes. os-architect and os-evolution-planner read this file to select the right delegation backend and cheapest brainstorm model automatically. Invoked by os-architect on first run if environment.md is absent.---## Roleos-environment-probe asks the user which AI environments they have access to, thenverifies each claimed environment by running a lightweight probe command. Results arewritten to `context/memory/environment.md` — a single source of truth that downstreamskills read to make delegation decisions without asking the user again.---## Environment InterviewAsk the user these questions (one prompt, multiple-choice, keep it brief):> Which of these AI tools do you currently have active on this machine?> (Select all that apply)>> A. Claude Code only> B. Claude Code + GitHub Copilot CLI (Pro or Business plan)> C. Claude Code + Agy CLI (Antigravity — gemini-3.5-flash backend)> D. Cursor (Claude or GPT backend)> E. Other (describe)Wait for their answer before probing.---## Probe CommandsFor each claimed environment, verify it is actually callable:| Environment | Probe command | Pass condition ||-------------|---------------|----------------|| Copilot CLI | `gh copilot explain "test" 2>&1 \| head -3` | No "not authenticated" or "command not found" || Agy CLI | `agy --version 2>&1 \| head -1` | Outputs a version string || Cursor | `cursor --version 2>&1 \| head -1` | Outputs a version string || Claude Code | always present | — |Report each probe result to the user:- Pass: "✓ Copilot CLI — confirmed"- Fail: "✗ Copilot CLI — not found or not authenticated (skipping)"Only write environments that pass to the profile.---## Procedure1. Run the Environment Interview.2. Run the Probe Commands for each claimed environment; only environments that pass get written.3. Write `context/memory/environment.md` in the format in `references/detailed-reference.md`.4. Downstream skills read this file: **os-evolution-planner** picks its brainstorm backend (Copilot CLI → Agy CLI → Claude Haiku subagent, in that priority order); **os-architect** picks its dispatch backend for Path B/C executions. If `environment.md` is missing, both default to Claude-only mode and offer to run this skill first.5. **Re-probe**: running this skill again overwrites `context/memory/environment.md` with fresh results — safe to re-run whenever the user's available environments change.Output format, smoke tests, gotchas, and support-file details are in`references/detailed-reference.md`.

Frequently asked questions