Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

WOZCODE Savings Benchmark

Runs a head-to-head benchmark of WOZCODE vs vanilla Claude Code on your own repo and reports cost, turn, and time savings.

Dev & CodingIntermediate★ 201⑂ 24AI score 8/10Last updated: Aug 5, 2026

What it does

  • Fires on phrases like "compare woz", "benchmark woz", "woz vs claude", "how much does woz save", or the /woz-benchmark command.
  • Validates the target first: directory exists, is a git repo, and has a clean working tree — otherwise it stops with fix instructions.
  • Collects 2–10 real coding tasks (plus optional environment setup commands) in a single short prompt and writes a temporary YAML config to /tmp/woz-benchmark-<timestamp>.yaml, defaulting to the opus model.
  • Executes benchmark.js with --user-env so each prompt runs twice (WOZCODE and vanilla), with git reset --hard restoring the repo between runs.
  • Relays the full report and prepends a savings summary computed from the totals.

Who it's for

  • Existing WOZCODE plugin users who are logged in.
  • Teams that need hard numbers (cost, turns, wall time) to justify a tool decision.
  • Developers who want agent performance measured on their actual codebase rather than synthetic tasks.

Example uses

  1. "how much does woz save?" → hand it a repo path and three refactor/bugfix tasks, get a comparison report.
  2. Run /woz-benchmark and include DB seed commands so data-dependent projects are benchmarked fairly.
  3. Point it at a repo with uncommitted work — it halts immediately and tells you to commit or stash first.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/WithWoz/wozcode-plugin/HEAD/codex/wozcode/skills/woz-benchmark/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the codex/wozcode/skills/woz-benchmark folder from the GitHub repo WithWoz/wozcode-plugin into my ~/.claude/skills/woz-benchmark/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/WithWoz/wozcode-plugin.git && mkdir -p ~/.claude/skills && cp -r wozcode-plugin/codex/wozcode/skills/woz-benchmark ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repo: git clone https://github.com/WithWoz/wozcode-plugin.git
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r wozcode-plugin/codex/wozcode/skills/woz-benchmark ~/.claude/skills/
  5. This skill depends on the WOZCODE plugin's benchmark.js. Install the plugin at ~/.codex/plugins/wozcode (or $CODEX_HOME/plugins/wozcode) per the repo README and log in to WOZCODE.
  6. Restart Claude Code, then type "benchmark woz" or /woz-benchmark.
  7. Make sure your benchmark target is a clean git repo — check with git status before starting, since runs use git reset --hard.
View source on GitHub ↗