grok

A set of shared instructions for the Grok coding-agent command-line tool. It describes where Grok finds settings, skills, plugins, and helper agents.

In plain words
What is it for?
Use it when configuring Grok, syncing skills, checking its setup, or working with its project-level integrations.
Why use it?
It helps keep Grok's project and personal configuration in sync and makes its limits and required environment settings clear.

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/wyattowalsh/agents/grok
Clone the repo
git clone --depth 1 https://github.com/wyattowalsh/agents

Made for: Cursor.

Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 1,246 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 $0.00000 $0.01246
Opus 5 $0.00000 $0.00623
Sonnet 5 $0.00000 $0.00249
Haiku 4.5 $0.00000 $0.00125

Measured 2d ago against content hash 8dcf0e58d25e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

grok 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.

.cursor/rules/grok.mdc · 95 lines

How it starts

The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Grok Build Global Instructions

Grok-Specific Overrides

Config surfaces

  • Home policy: ~/.grok/config.toml — models, features, memory, subagents, compat layers, plugins.
  • Project MCP: .grok/config.toml — MCP servers only (plus optional plugins/permission).
  • Repo source: config/grok-config.toml merged by uv run python scripts/sync_agent_stack.py --apply --platforms grok.

Skills discovery

Grok reads ~/.grok/skills/, project .grok/skills/, repo plugin skills, and Claude-compat ~/.claude/skills/. Skills CLI has no native grok adapter; use wagents install -a grok or wagents skills sync -a grok (installs via claude-code, then mirrors to ~/.grok/skills).

Subagents

Grok hard-caps nested subagent depth at 1. Do not invent deeper orchestration knobs in repo config.

Experimental env vars

Some toggles are env-only. Source config/grok-env.sh in your shell:

source /path/to/agents/config/grok-env.sh

Required exports: GROK_WEB_FETCH, GROK_MEMORY, GROK_SUBAGENTS, GROK_LSP_TOOLS. Run uv run wagents grok doctor --format json to verify.

This repo expects a local direnv file (.envrc is gitignored) with source_env "${PWD}/config/grok-env.sh". Grok policy sets [session] load_envrc = true so that file loads when you work inside the clone. Without direnv, run source config/grok-env.sh manually.

LSP (global)

Repo SSOT: config/grok-lsp.json — synced to ~/.grok/lsp.json on home sync (all projects inherit unless a project overrides).

Server Strategy Install hint
typescript npx -y typescript-language-server npm i -g typescript-language-server typescript
python npx -y pyright-langserver npm i -g pyright
json / yaml / toml / bash npx -y wrappers Usually no global install needed
markdown marksman server (PATH) brew install marksman
rust rust-analyzer (PATH) rustup component add rust-analyzer
go gopls (PATH) go install golang.org/x/tools/gopls@latest

Read the full file on GitHub · 95 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 · 95 lines · 1,246 tokens per session scan A 8dcf0e58d25e

Subscribe to this mod's changes

grok is a cursor rule published in the GitHub repository wyattowalsh/agents (5 stars, last pushed 11d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,246 tokens. 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.