A disposable testing workflow that checks every page using a content block on a chosen Git branch and captures screenshots with a local development server.

Use it for a visual sweep of a block on the stage branch or another existing branch, with optional screenshots of the full page or a selected element.

It lets you inspect a branch's visual result without changing your main checkout or manually opening each affected page.

How adobecom/da-express-milo configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to adobecom/da-express-milo. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/adobecom/da-express-milo/stage/.claude/skills/qa-worktree/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/adobecom/da-express-milo

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/adobecom/da-express-milo/qa-worktree/github.svg)](https://agentmods.dev/skills/adobecom/da-express-milo/qa-worktree)

Your own site

<a href="https://agentmods.dev/skills/adobecom/da-express-milo/qa-worktree"><img src="https://agentmods.dev/badge/skills/adobecom/da-express-milo/qa-worktree/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/adobecom/da-express-milo/qa-worktree"><img src="https://agentmods.dev/badge/skills/adobecom/da-express-milo/qa-worktree.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 144 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,400 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00144	$0.01400
Opus 5.5	$0.00058	$0.00560
Sonnet 5.5	$0.00029	$0.00280
Haiku 4.5	$0.00014	$0.00140

Measured 1mo ago against content hash adac741cd806, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

qa-worktree 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 1mo 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.

.claude/skills/qa-worktree/SKILL.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

QA Worktree Skill

Backed by .claude/tools/qa-worktree/qa-worktree.mjs (lives in this repo, not globally), which is independently invokable by any skill/agent.

Workflow

Get the block name from the invocation args (e.g. /qa-worktree grid-marquee). If nothing was passed, ask for one.
Get the ref (optional). Default is stage. If the user names a branch ("QA this on my branch", "check it on "), pass that as --ref. The branch must already exist on origin — this tool checks out origin/<ref> into a scratch worktree; it does not push or create remote branches.
Run the tool (resolve the repo root first — this tool lives inside the checkout, so don't assume cwd is the repo root):
```
ROOT="$(git rev-parse --show-toplevel)"
node "$ROOT/.claude/tools/qa-worktree/qa-worktree.mjs" --block="<name>" --ref="<ref>"
```
Options worth knowing about:
- --selector="<css>" — capture an element crop in addition to the full page (default mode becomes both once a selector is given). Useful when the page has a lot of noisy unrelated content.
- --mode=<full|element|both> — force a specific mode. Default is full unless --selector is given. Use --mode=element to skip the full-page screenshot entirely for speed (this is what /comparison does by default, since block comparisons rarely need the whole page).
- --concurrency=<n> — pages captured in parallel (each in its own browser context). Default 6. Bump it for a large affected-page count if the machine can take it; drop it if aem up starts choking.
- --timeout=<seconds> — dev-server readiness window. Default 60; raise it on a slower machine or a cold first aem up.
- --locale=<key> (repeatable) / --all-locales — QA other locales, not just en (the default). A locale requested explicitly via --locale that has no local content checkout is an error; --all-locales skips it instead (see localesSkipped in the output) and syncing locales is a separate step — node sync-locale-content.mjs --list shows what's synced, --locale=<key>/--all syncs more.

Read the full file on GitHub · 101 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.

1mo ago First seen · 101 lines · 144 tokens per session scan A adac741cd806

Subscribe to this mod's changes

qa-worktree is a skill published in the GitHub repository adobecom/da-express-milo (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 144 tokens to every session and 1,400 once invoked, about $0.0006 per session on Opus 5.5. 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 2026-09-10.

CSS and styling Frontend Git and workflow

qa-worktree next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
qa-worktreeadobecom/da-express-milo	Skill	6	yesterday	Apache-2.0	A	144	Claude Code
ui-ux-architectralvarezdev/ralvaskills	Skill	2	7d ago	GPL-3.0	A	77	Claude Code, Codex
accesslint-diffLiHongwei-cn/lihongwei-cn	Skill	5	29d ago	MIT	A	68	Claude Code
visual-ralphYeachan-Heo/oh-my-codex	Skill	34k	3d ago	MIT	A	52	Codex
prowler-commitprowler-cloud/prowler	Skill	15k	yesterday	Apache-2.0	A	44	Claude Code, Codex
gh-auth-isolationgithub/gh-aw	Skill	5.6k	yesterday	MIT	A	20	Claude Code, Codex
prototype-webnexu-io/html-anything	Skill	9.4k	25d ago	Apache-2.0	A	24	Claude Code, Codex

da-express-milo: Skill for Claude Code