global-agent-instructions

A set of instructions that applies to coding agents across the machine. It is read from a shared AGENTS.md file so different agent tools follow the same rules.

In plain words
What is it for?
Use it to define rules for agent behavior, such as formatting requirements and restrictions on commit authorship.
Why use it?
It prevents agents from ignoring important personal or project-wide policies. The live file is treated as the source of truth when its contents change.

Cursor rule for Cursor

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 rules/ctt062/dotfiles/global-agent-instructions
Clone the repo
git clone --depth 1 https://github.com/ctt062/dotfiles

Made for: Cursor.

Per session 900 This file is loaded in full into every session.
When invoked 900 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 $0.00900 $0.00900
Opus 5 $0.00450 $0.00450
Sonnet 5 $0.00180 $0.00180
Haiku 4.5 $0.00090 $0.00090

Measured yesterday against content hash 12e961a715cc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

global-agent-instructions 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 yesterday.

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.

home/.cursor/plugins/global-agents/rules/global-agent-instructions.mdc · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Mandatory first step

Before any other tools or answers, read ~/.cursor/AGENTS.md with the Read tool and follow it for the rest of this conversation.

That path is a symlink to ~/.dotfiles/home/AGENTS.md, the single source of truth for global agent policy on this machine. If the live file differs from the snapshot below, the live file wins.

Snapshot of home/AGENTS.md

Global agent instructions

Read and follow this file before doing any work. This is mandatory for every agent on this machine (Claude Code, Codex, Cursor, opencode, Grok, and any other agent harness).

home.nix installs this file for Claude (~/.claude/CLAUDE.md), Codex (~/.codex/AGENTS.md), Cursor (~/.cursor/AGENTS.md), opencode (~/.config/opencode/AGENTS.md), and Grok (~/.grok/AGENTS.md).

Rules

  • Never use the em dash "—". Use plain dash "-" instead.
  • Never add yourself (or any AI / agent) as a git co-author unless the user explicitly asks for it. This applies to Cursor, Claude, Codex, opencode, Grok, and every other agent identity.
    • Do not pass --trailer "Co-authored-by: ..." (or any Co-authored-by / Co-Authored-By text) to git commit.
    • Forbidden example: git commit --trailer "Co-authored-by: Cursor <[email protected]>".
    • Commit with only -m / HEREDOC message body. No agent attribution trailers in commits or PR bodies.
  • Never manually modify CHANGELOG.md files or any files marked as auto-generated.
  • When making technical decisions, do not give much weight to development cost. Prefer quality, simplicity, robustness, scalability, and long-term maintainability.
  • When fixing bugs, always start by reproducing the bug in an E2E setting as close as possible to how an end user would experience it, so the fix addresses the real problem.
  • When end-to-end testing a product, be picky about the UI and obsess over pixel perfection. If something clearly looks off, even if it is unrelated to the current task, fix it along the way.
  • Apply the same standard to engineering excellence: lint failures, test failures, and flakiness. If you see one, even if it is unrelated to the current task, fix it.
  • Prefer AXI over MCP and raw vendor CLIs for the same job:
    • GitHub: gh-axi (not gh, not GitHub MCP)
    • Visual review artifacts: lavish-axi
    • Validation / shipping gates: no-mistakes axi ... If an AXI skill is available, follow that skill.
  • Before using "dynamic workflows", "ultra code", or any harness feature that immediately spawns a large swarm of subagents, explain the tradeoffs and ask the user for explicit approval.

Read the full file on GitHub · 53 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. yesterday First seen · 53 lines · 900 tokens per session scan A 12e961a715cc

Subscribe to this mod's changes

global-agent-instructions is a cursor rule published in the GitHub repository ctt062/dotfiles (2 stars, last pushed 21d ago), licensed MIT. It adds 900 tokens to every session, about $0.0045 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 2026-08-31.