go

A command that starts an autonomous work loop for a business project. It repeatedly runs short work cycles that ship something, learn a fact, record it, and push the changes.

In plain words
What is it for?
Use it to start recurring work cycles, set a time cadence such as 30m or 1h, and pass a focus area to the loop.
Why use it?
It turns repeated progress steps into an ongoing loop until someone stops it.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/patwalls/headroom/go
Clone the repo
git clone --depth 1 https://github.com/patwalls/headroom

Made for: Claude Code.

Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 664 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00029 $0.00664
Opus 5 $0.00015 $0.00332
Sonnet 5 $0.00006 $0.00133
Haiku 4.5 $0.00003 $0.00066

Measured 3d ago against content hash b9f406c88005, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

go scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/go.md · 39 lines

How it starts

The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/go — ignition

One command = the whole machine. /go starts this venture's autonomous entrepreneur loop: run /lap (ship one real thing toward the North Star + learn one fact, log it, push) on repeat until told to stop. It runs /lap on repeat as fresh, token-frugal one-shots (no day-long context).

What to do

  1. Sync + label the session (the equivalent of /rename HEADROOM LOOP — puts the business name in the cmux sidebar + session list). Run these; if either fails, continue — cosmetic:

    git pull --rebase --autostash   # Pat ships from other machines — start from latest
    [ -n "$CMUX_CLAUDE_HOOK_CMUX_BIN" ] && CMUX_QUIET=1 "$CMUX_CLAUDE_HOOK_CMUX_BIN" rename-workspace "HEADROOM LOOP" || true
    node -e 'const fs=require("fs"),os=require("os"),d=os.homedir()+"/.claude/sessions";const id=process.env.CLAUDE_CODE_SESSION_ID;for(const f of fs.readdirSync(d)){const p=d+"/"+f;try{const j=JSON.parse(fs.readFileSync(p,"utf8"));if(j.sessionId===id){j.name="HEADROOM LOOP";fs.writeFileSync(p,JSON.stringify(j))}}catch{}}'
    
  2. Parse $ARGUMENTS — a \d+[smhd] token (e.g. 1h, 30m) is the cadence (default 20m if absent); anything else is a focus string handed to every lap.

  3. Launch the fresh-context runner, detached. Each lap is a brand-new claude -p "/lap" with ZERO inherited context — token-frugal, it never re-reads a growing conversation. Run, substituting the parsed cadence + focus (drop the focus if none):

    nohup ~/.claude/loop-runner.sh <cadence> <focus> >> "$PWD/.loop.log" 2>&1 &
    echo "loop running detached — PID $!"
    

    The runner refuses to start if one is already running for this repo (a .loop.pid lock).

  4. Tell Pat it's live: the cadence, that it KEEPS RUNNING after this session closes (it's detached), how to watch (tail -f .loop.log), and how to control it — /throttle-loops <interval> to slow it, /pause-loops to stop everything. Do NOT run a lap yourself in this session; the runner owns laps now.

Read the full file on GitHub · 39 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 39 lines · 29 tokens per session scan A b9f406c88005

Subscribe to this mod's changes

go is a command published in the GitHub repository patwalls/headroom (10 stars, last pushed 2mo ago), licensed MIT. It adds 29 tokens to every session and 664 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.