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.
npx agentmods add skills/justrach/codegraff2/create-commandnpx skills add justrach/codegraff2 --skill create-commandgit clone --depth 1 https://github.com/justrach/codegraff2What 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 | $0.00082 | $0.03811 |
| Opus 5 | $0.00041 | $0.01906 |
| Sonnet 5 | $0.00016 | $0.00762 |
| Haiku 4.5 | $0.00008 | $0.00381 |
Grade C, and why
create-command scanned grade C 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 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
<shell>rm -rf target/debug</shell> This is a copy
100% identical to create-command — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 710 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create Commands
Create and manage commands for the code-forge application. Commands are modular workflows that can be invoked to perform specific tasks.
File Location
CRITICAL: All command files must be created in the <cwd>/.forge/commands directory, where <cwd> is the current working directory of your code-forge project.
- Directory:
<cwd>/.forge/commands - File format:
{command-name}.md - Example: If your project is at
/home/user/my-project, commands go in/home/user/my-project/.forge/commands/
This is the only location where forge will discover and load custom commands.
Command File Structure
Every command file must have:
-
YAML Frontmatter (required):
name: Command identifier (use hyphens for multi-word names)description: What the command does
-
Command Body (required):
- List of steps to execute
- Special tags for automated workflows
- Clear instructions for each step
Example Command File
---
name: check
description: Checks if the code is ready to be committed
---
- Run the `lint` and `test` commands and verify if everything is fine.
<lint>cargo +nightly fmt --all; cargo +nightly clippy --fix --allow-staged --allow-dirty --workspace</lint>
<test>cargo insta test --accept --unreferenced=delete</test>
- Fix every issue found in the process
Complete Sample Command
This sample demonstrates all three tag types:
---
name: sample-command
description: Sample command demonstrating the command file structure
---
This is a sample command that demonstrates the structure of command files.
- First step: Perform an initial action
<lint>echo "Running linting..."</lint>
- Second step: Execute tests
<test>echo "Running tests..."</test>
- Third step: Complete the workflow
<shell>echo "Workflow complete!"</shell>
- Final step: Verify everything worked correctly
Creating a New Command
Step 1: Determine Command Purpose
Identify what the command should accomplish:
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.
- yesterday First seen · 710 lines · 82 tokens per session scan C 18e7e101b3a8
create-command is a skill published in the GitHub repository justrach/codegraff2 (24 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 82 tokens to every session and 3,811 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). It is 100% identical to create-command, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…