code-airlock AGENTS.md

A set of project instructions for working on code-airlock, a Bash tool that runs coding agents inside disposable Docker-based virtual machines.

In plain words
What is it for?
Use it when editing code-airlock, checking its Bash script, updating its README, or moving completed changes out of a sandbox.
Why use it?
It gives contributors shared rules for safe changes, security settings, documentation, verification, and retrieving work from the isolated environment.

Instructions file for CodexOpenCode

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 instructions/trivo25/code-airlock/agents-md
Clone the repo
git clone --depth 1 https://github.com/Trivo25/code-airlock

Made for: Codex, OpenCode.

Per session 278 This file is loaded in full into every session.
When invoked 278 The same file — it is already loaded in full.
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 $0.00278 $0.00278
Opus 5 $0.00139 $0.00139
Sonnet 5 $0.00056 $0.00056
Haiku 4.5 $0.00028 $0.00028

Measured 2d ago against content hash 7d17be5fe803, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-airlock AGENTS.md 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 2d 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.

AGENTS.md · 40 lines

What it actually says

Agent Instructions

This repo is code-airlock: a small Bash wrapper around Docker Sandboxes for running coding agents in disposable microVMs.

Working Rules

  • Keep the tool dependency-light. Do not add package managers, build systems, or new runtime dependencies unless explicitly requested.
  • Prefer small, reviewable Bash changes in code-airlock.
  • Keep docs in README.md aligned with script behavior.
  • Do not change security defaults casually. Explain any change to network policy, credential handling, clone mode, or sandbox deletion behavior.
  • Do not push to GitHub or publish releases unless explicitly asked.

Verification

After editing the wrapper, run:

bash -n code-airlock
shellcheck code-airlock
./code-airlock help

For docs-only changes, at least check for stale command names:

rg -n "claude-sbx|code-airlock|sbx " README.md code-airlock sandbox.conf.example AGENTS.md

Sandbox Workflow

The agent works in Docker Sandboxes clone mode. Commit any finished edits inside the sandbox so the host can retrieve them with:

code-airlock fetch
code-airlock diff
code-airlock merge

If work is not committed in the sandbox, it will not come back through the normal fetch/diff/merge workflow.

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. 2d ago First seen · 40 lines · 278 tokens per session scan A 7d17be5fe803

Subscribe to this mod's changes

code-airlock AGENTS.md is an instructions file published in the GitHub repository Trivo25/code-airlock (20 stars, last pushed 1mo ago), licensed MIT. It adds 278 tokens to every session, about $0.0014 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.