hal-9000 AGENTS.md

A project instruction file for hal-9000, a macOS automation and developer-tools repository. It explains how to use the repository's commands and how its configuration and skills are distributed.

In plain words
What is it for?
It helps maintain dotfiles, Python and Ansible checks, Claude Code skills, plugin metadata, and the hal command-line tools.
Why use it?
It keeps agents from editing mirrored files directly, running the wrong checks, or updating only one of several intentionally duplicated configuration records.

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/vinta/hal-9000/agents-md
Clone the repo
git clone --depth 1 https://github.com/vinta/hal-9000

Made for: Codex, OpenCode.

Per session 786 This file is loaded in full into every session.
When invoked 786 The same file — it is already loaded in full.
Security scan B 1 finding. 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 $0.00786 $0.00786
Opus 5 $0.00393 $0.00393
Sonnet 5 $0.00157 $0.00157
Haiku 4.5 $0.00079 $0.00079

Measured 2d ago against content hash 4665a4b97ca1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

hal-9000 AGENTS.md scanned grade B with 1 finding 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- **Skills are the source of truth in `skills/`**: Distributed via Claude Code plugin marketplaces configured in `dotfiles/.claude/settings.json` (the `hal-9000` marketplace loads the published version from GitHub), and
AGENTS.md · 50 lines

How it starts

The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

macOS dev environment automation: dotfiles, AI agent configs, skills, and dev stacks.

Commands

Run make help to list targets and hal --help for the CLI. Use make targets instead of running the underlying commands directly. They chain the right tools with the right flags (e.g. make lint-python runs ruff format --check and ruff check; make lint-ansible runs ansible-lint and a playbook syntax check; make lint runs both).

Gotchas

  • Dotfiles are the source of truth: dotfiles/ is the source of truth for files under ~/. dotfiles/.claude/ syncs to ~/.claude/ via hal_dotfiles.json. Always edit under dotfiles/, never under ~/ directly.
  • Skills are the source of truth in skills/: Distributed via Claude Code plugin marketplaces configured in dotfiles/.claude/settings.json (the hal-9000 marketplace loads the published version from GitHub), and via npx skills add vinta/hal-9000.
  • The hal-skills skills array is duplicated on purpose: skills/.claude-plugin/plugin.json is what Claude Code reads for the plugin's author and skill count, and the copy in .claude-plugin/marketplace.json is what npx skills reads to group the skills under "Hal Skills" (it only looks for a plugin.json at the repo root, so it never sees the one under skills/). Keep both lists in sync instead of deduplicating them.
  • For generated artifacts such as zsh completion, regenerate them with the repo command instead of editing them by hand (e.g. make hal-completion after modifying bin/hal.py).

External Tool Documentation

When you need information about tools used in this project, use the find-docs skill or WebFetch.

Context7 Library IDs

Pre-resolved IDs for the find-docs skill. Pass directly to ctx7 docs, skipping the ctx7 library step:

Tool libraryId
ansible /websites/ansible_projects_ansible
ansible-lint /ansible/ansible-lint
betterleaks /betterleaks/betterleaks
fnm /schniz/fnm
github-actions /websites/github_en_actions
homebrew /homebrew/brew
oh-my-zsh /ohmyzsh/ohmyzsh
ollama /ollama/ollama
pre-commit /pre-commit/pre-commit.com
pytest /pytest-dev/pytest
ruff /websites/astral_sh_ruff
ty /websites/astral_sh_ty
uv /websites/astral_sh_uv
zsh /websites/zsh_sourceforge_io_doc_release

Read the full file on GitHub · 50 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. 2d ago First seen · 50 lines · 786 tokens per session scan B 4665a4b97ca1

Subscribe to this mod's changes

hal-9000 AGENTS.md is an instructions file published in the GitHub repository vinta/hal-9000 (125 stars, last pushed 5d ago), licensed MIT. It adds 786 tokens to every session, about $0.0039 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.