Borrowing it
Nothing to install: this file belongs to a-b/readable-ad-hoc-commands. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/a-b/readable-ad-hoc-commands/main/AGENTS.mdgit clone --depth 1 https://github.com/a-b/readable-ad-hoc-commandsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/a-b/readable-ad-hoc-commands/agents-md)<a href="https://agentmods.dev/instructions/a-b/readable-ad-hoc-commands/agents-md"><img src="https://agentmods.dev/badge/instructions/a-b/readable-ad-hoc-commands/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/a-b/readable-ad-hoc-commands/agents-md"><img src="https://agentmods.dev/badge/instructions/a-b/readable-ad-hoc-commands/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01909 | $0.01909 |
| Opus 5.5 | $0.00764 | $0.00764 |
| Sonnet 5.5 | $0.00382 | $0.00382 |
| Haiku 4.5 | $0.00191 | $0.00191 |
Grade A, and why
readable-ad-hoc-commands 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 2mo 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.
How it starts
The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Readable ad-hoc commands — agent instructions
These are drop-in instructions for any coding agent. Put them in a system
prompt, an AGENTS.md, a Cursor/Windsurf rules file, or load the bundled Claude
Code skill (see readable-ad-hoc-commands/SKILL.md).
The goal: when an agent runs a non-trivial command, format it so the human can
grasp its intent, steps, and blast radius before it executes — instead of a
dense one-liner they must reverse-engineer or approve blind.
This applies to any command language you fire at a terminal or REPL, not just
bash — python -c, node -e, jq, psql -c, awk. It is about throwaway
commands, not maintained application code; writing clean application code is a
different concern (that's code review's job).
When this applies
Apply the rules to any command that is not a trivial single invocation:
- Multi-line scripts, or
&&/;/|chains of 2+ meaningful steps - Loops, conditionals, functions,
xargs - Here-docs (
<<EOF) and multi-line string literals - Dense inline snippets:
python -c "...",node -e "...",jq '...',psql -c "..." - Non-obvious flags, subshells, process substitution, complex quoting/escaping
- Anything that writes, deletes, moves, or overwrites files, tables, or state
A bare ls, git status, or SELECT 1 needs none of this — just run it.
The rules
These are language-neutral. The mechanics (comment char, fail-fast idiom) differ per language — see the table below.
- One step per line, top to bottom. Break dense one-liners into a readable
sequence. The reader should be able to scan line-by-line and follow the logic.
Prefer newlines over
;. For inline snippets, use a here-doc or a real multi-line string instead of cramming everything onto one-cline. - Comment the why, not the obvious. Add a short comment above each logical block explaining its purpose and, if relevant, its effect. Skip comments that just restate the command.
- Name things. Pull magic values, paths, and repeated strings into named variables/bindings at the top. A named value is self-documenting.
- Announce progress for long runs. For multi-phase work, print a short banner before each phase so the user can follow along in the output, not just the source.
- Fail loud and early. Start real scripts with the language's fail-fast idiom so a failed step stops the run instead of silently barreling on. Mention it exists; don't over-explain.
- Preview destructive actions. Before anything that deletes/overwrites, print exactly what will be affected, or do a dry run first (or wrap it in a transaction you can roll back). Never bury a destructive step mid-pipeline.
- Say what it does in prose, first. Before running, give the user a one- to three-line plain-English summary of what the command does and what changes to expect. Then show the command.
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.
- 2mo ago First seen · 180 lines · 1,909 tokens per session scan A 51a61085ea24
readable-ad-hoc-commands AGENTS.md is an instructions file published in the GitHub repository a-b/readable-ad-hoc-commands (2 stars, last pushed 3mo ago), licensed MIT. It adds 1,909 tokens to every session, about $0.0076 per session on Opus 5.5. 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 2026-08-31.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.