← All skills
orch
97 / 100 · Low risk
Fine to try.
By lev-os/agents (GitHub) · research · 27 stars · scanned at commit 201a805
Use when coordinating a multi-step goal across agents, skills, Lev lifecycle routes, scheduling, research, implementation, review, or human decisions.
Works with any agent that supports SKILL.md skills.
What this skill can do
Runs commands on your computer
This is disclosed by the author in the skill's SKILL.md or README.
Found in 44 places.
Show 17 more
- skills/orch/evals/runs/e002/run_trials.py, line 165
checks['independent_regression_pass'] = subprocess.run([sys.executable, '-c', SLUG_CHECK], cwd=root, capture_output=True, timeout=20).returncode == 0 - skills/orch/evals/runs/e002/run_trials.py, line 215
process = subprocess.Popen(command, cwd=root, env=env, stdin=subprocess.PIPE, - skills/orch/evals/runs/e002/run_trials.py, line 221
except subprocess.TimeoutExpired: - skills/orch/evals/runs/e002/run_trials.py, line 225
except subprocess.TimeoutExpired: - skills/orch/evals/runs/e002/sample_host.py, line 35
proc = subprocess.run(item['command'], cwd=ROOT, text=True, capture_output=True) - skills/orch/evals/runs/e002/sample_host.py, line 39
proc = subprocess.run(item['verifier'], cwd=ROOT, text=True, capture_output=True) - skills/orch/evals/runs/e002/test_agent_grader.py, line 25
result = subprocess.run([sys.executable, 'host.py', *args], cwd=root, text=True, capture_output=True) - skills/orch/evals/runs/e003/run_trials.py, line 218
process = subprocess.Popen(command, cwd=root, env=env, stdin=subprocess.PIPE, - skills/orch/evals/runs/e003/run_trials.py, line 224
except subprocess.TimeoutExpired: - skills/orch/evals/runs/e003/run_trials.py, line 228
except subprocess.TimeoutExpired: - skills/orch/evals/runs/e003/sample_host.py, line 26
return subprocess.run(['/usr/bin/sandbox-exec', '-p', profile, interpreter, *command[1:]], - skills/orch/evals/runs/e003/test_agent_grader.py, line 25
result = subprocess.run([sys.executable, 'host.py', *args], cwd=root, text=True, capture_output=True) - skills/orch/evals/runs/e003/test_trial_boundary.py, line 40
result = self.run_code("import subprocess; subprocess.run(['/usr/bin/true'], check=True)") - skills/orch/evals/runs/e003/test_trial_boundary.py, line 55
result = subprocess.run([sys.executable, str(Path(__file__).with_name('sample_mcp.py')), str(self.root)], - skills/orch/evals/runs/e005/run_trials.py, line 243
process = subprocess.Popen(command, cwd=root, env=env, stdin=subprocess.PIPE, - skills/orch/evals/runs/e005/run_trials.py, line 249
except subprocess.TimeoutExpired: - skills/orch/evals/runs/e005/run_trials.py, line 253
except subprocess.TimeoutExpired:
and 24 more places
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'orch' ] || [ -L ~/.agents/skills/'orch' ]; then echo 'A skill folder named orch already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/lev-os/agents' "$d" && git -C "$d" checkout --quiet '201a80515e1dc58addfda3cd9978907de2c944a6' -- 'skills/orch' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/orch' ~/.agents/skills/'orch' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'orch' ~/.claude/skills/'orch' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
The badge for lev-os/agents shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/lev-os/agents/)
Score history
| Date | Version | Score |
|---|
| 2026-11-07 | 201a805 | 97 |