Install the "arbitrage" agent skill from https://github.com/blader/arbitrage/tree/main into .claude/skills/arbitrage/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "arbitrage", then confirm the skill loads.
Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add blader/arbitrage --skill arbitrage -a codex
Project install goes to .agents/skills/; add -g for ~/.codex/skills/.
Install the "arbitrage" agent skill from https://github.com/blader/arbitrage/tree/main into .agents/skills/arbitrage/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "arbitrage", then confirm the skill loads.
Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add blader/arbitrage --skill arbitrage -a cursor
Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.
Install the "arbitrage" agent skill from https://github.com/blader/arbitrage/tree/main into .cursor/skills/arbitrage/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "arbitrage", then confirm the skill loads.
Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add blader/arbitrage --skill arbitrage -a gemini-cli
Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.
Install the "arbitrage" agent skill from https://github.com/blader/arbitrage/tree/main into .gemini/skills/arbitrage/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "arbitrage", then confirm the skill loads.
Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
GitHub CLI
$ gh skill install blader/arbitrage arbitrage
Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
skills CLI
$ npx skills add blader/arbitrage --skill arbitrage -a github-copilot
Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.
Install the "arbitrage" agent skill from https://github.com/blader/arbitrage/tree/main into .github/skills/arbitrage/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "arbitrage", then confirm the skill loads.
GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add blader/arbitrage --skill arbitrage -a opencode
OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
Install the "arbitrage" agent skill from https://github.com/blader/arbitrage/tree/main into .opencode/skills/arbitrage/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "arbitrage", then confirm the skill loads.
OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
Skill name
arbitrage
GitHub stars
175
Token cost
~1.1k tokens
SKILL.md length
586 words
Files
3
Skills in repo
1
Repo updated
First seen
Licence
MIT
At a glance
Always active when coding in a Claude (Fable) session. An agent skill from blader/arbitrage.
Works in 3 steps: Spec first. Write a spec file in the… → Dispatch in the background and keep… → Review the diff here. Commit, push, and…
Ever implementation work is being planned
SKILL.md covers Routing Table, Dispatch Protocol, Visual Validation Loop… and Escape Hatch: When codex…, plus 2 more sections
Calls codex
What it does
Arbitrage is an agent skill from blader/arbitrage. Always active when coding in a Claude (Fable) session. Triggers whenever implementation work is being planned, scoped, or about to start — before writing any code — to decide where each piece of work executes.
Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `README.md`).
The repository describes itself as: A Claude Code skill for token arbitrage: keep premium model tokens for judgment, dispatch all code-writing to codex. The licence is MIT.
When your agent uses it
Ever implementation work is being planned
About to start — before writing any code — to decide where each piece of work executes
Example prompts
“Use the arbitrage skill to alway active when coding in a Claude (Fable) session. An agent skill from blader/arbitrage”
“/arbitrage”
Workflow steps
3 steps, taken from the first numbered list in SKILL.md.
1Spec first. Write a spec file in the worktree: objective, constraints, files/areas involved, acceptance criteria — the exact test command…
2Dispatch in the background and keep working
3Review the diff here. Commit, push, and open PRs from this session.
What it can do on your machine
Read from SKILL.md and the folder at commit ccfd550. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
codex
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Arbitrage loads about 1.1k tokens when it runs. Until then it costs about 55 tokens; SKILL.md has 586 words of instructions outside code blocks.
Always· name and description, kept in context so the agent knows when to use it
~55
When it runs· the whole SKILL.md, loaded when a task matches
~1.1k
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
Download SKILL.mdSave it as .claude/skills/arbitrage/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
arbitrage
description
Always active when coding in a Claude (Fable) session. Triggers whenever implementation work is being planned, scoped, or about to start — before writing any code — to decide where each piece of work executes.
Arbitrage
Fable tokens are expensive; the user's codex quota sits unused. Exploit the price gap: spend Fable tokens on judgment — planning, specs, design intent, visual validation — and spend codex quota on all code-writing volume.
You ARE the Fable session. "Here" means you do it yourself. There is no separate Fable to delegate to.
Routing Table
Work
Where
Why
Planning, specs, architecture decisions
Here
Judgment is what Fable is paid for
ALL implementation — backend, frontend, anything that writes code
codex via /goal
Quota is free; typing code is not where Fable adds value
Visual validation of UI work (run the app, screenshot, judge, iterate)
Here
Fable's design eye is worth the price — validate, don't type
Investigation and debugging analysis (root-causing hard bugs)
Here
Judgment work; the resulting fix dispatches with a precise spec
Diff review, commits, pushes, PRs
Here
Git operations stay under your control
Trivial edits riding along with review/validation (one-liners, any layer)
Here
Dispatch overhead exceeds the work
Work codex has demonstrably struggled with
Here
Escape hatch — see below
Dispatch Protocol
Spec first. Write a spec file in the worktree: objective, constraints, files/areas involved, acceptance criteria — the exact test command that must go green — and what NOT to touch. For frontend work the spec carries the design intent: layout, states, interactions, spacing, motion, reference patterns. For the hardest parts, spec down to pseudocode — the thinking stays here; the typing still dispatches.
Dispatch in the background and keep working:bash
codex exec --cd <worktree> "/goal <one-line objective; details in SPEC.md>"
codex runs gpt-5.5 with xhigh reasoning (its default config); it implements and gets tests green in the isolated worktree while you continue planning/spec/validation work in parallel.
Review the diff here. Commit, push, and open PRs from this session.
Visual Validation Loop (frontend)
codex implements the UI per spec.
Run the app here — interact, screenshot, judge against the design intent.
Write concrete visual feedback (what's off, by how much, what good looks like) and re-dispatch.
Loop until it looks right; then diff review + git here.
Show full SKILL.md (256 more words)Show less
Escape Hatch: When codex Struggles
Struggle must be observed, not predicted. Always dispatch first — predicting "codex can't do this" just means the spec needs pseudocode-level detail. But two strikes and it's yours: if a re-dispatch with concrete corrective feedback still comes back missing the acceptance criteria or mangling the approach, stop re-dispatching and write the code here yourself, salvaging whatever of codex's diff is sound.
Red Flags
Thought
Reality
"Frontend is design-critical, so I'll code it here"
Fable's taste shows up in the spec and the validation loop, not in typing JSX. Dispatch it.
"Faster to just write it myself"
Premium tokens on rote work. Spec it, dispatch it, validate in parallel.
"This part is too hard for codex"
Predicted struggle is not observed struggle. Spec it at pseudocode level and dispatch; you only take over after two failed rounds.
"I'll let codex commit and push"
Review the diff and run git from this session.
"No time to write a spec"
An unspecced dispatch comes back wrong and costs more Fable tokens to fix than the spec would have.
Common Mistakes
Dispatching without acceptance criteria → codex returns "done" with red tests. Always name the test command that must pass.
Dispatching frontend work without design intent in the spec → generic UI comes back; the validation loop burns rounds recovering what the spec should have said.
Blocking on codex → run it in the background; keep doing planning/spec/validation work in parallel.
Splitting one coherent unit across dispatches → pick one owner per unit; split at the API boundary, not mid-feature.
Arbitrage next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
Summarize recent GitHub activity — discussions, PRs, issues, events, traffic — into an actionable report so you can stay on top of the project without reading everything.
Executes containment strategies to stop active adversary operations and prevent lateral movement during a confirmed security breach.
34k GitHub stars~2.7k tokensUpdated 1 mo ago
SecurityAuto-check passed
Questions about Arbitrage
What does Arbitrage do?
Always active when coding in a Claude (Fable) session. An agent skill from blader/arbitrage. Arbitrage is an agent skill from blader/arbitrage. Always active when coding in a Claude (Fable) session.
When should I use Arbitrage?
Arbitrage fits situations like: ever implementation work is being planned; about to start — before writing any code — to decide where each piece of work executes.
How do I install Arbitrage in Claude Code?
Run `npx skills add blader/arbitrage --skill arbitrage -a claude-code`. Or copy the skill folder (the blader/arbitrage repository) into .claude/skills/arbitrage in your project. Claude Code loads it when a task matches its description.
How do I install Arbitrage in Codex?
Run `npx skills add blader/arbitrage --skill arbitrage -a codex`. Or copy the skill folder (the blader/arbitrage repository) into .agents/skills/arbitrage in your project. Codex loads it when a task matches its description.
Can I use Arbitrage in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add blader/arbitrage --skill arbitrage -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/arbitrage, .gemini/skills/arbitrage, .github/skills/arbitrage and .opencode/skills/arbitrage in your project.
What does Arbitrage need to run?
Going by SKILL.md and its folder, Arbitrage needs the command-line tools its instructions call (codex).
Does Arbitrage access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Arbitrage safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Arbitrage use?
Arbitrage is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Arbitrage use?
About 1.1k tokens (SKILL.md is roughly 4.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Arbitrage?
Skills that share tags, products or a category with Arbitrage: Fable Fable (mrtooher/fable-mode, 870 stars), Summarize Activity (alpinejs/alpine, 32k stars), Arbitrage (alsk1992/CloddsBot, 2.9k stars) and Duplicate Id Active (thedaviddias/Front-End-Checklist, 74k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Arbitrage?
blader (a GitHub user) maintains it in blader/arbitrage, which has 175 GitHub stars. The repository was last updated on June 12, 2026.
Source: blader/arbitrage on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.
