Audit cloned or reimplemented websites for fidelity gaps, tracking scripts, source-brand and language residue, placeholders, and risky external dependencies. Use before handoff or deployment, or when asked to review a website clone for cleanup and readiness.
SkillScore65Maintained
Automatic
◆QA & Test Engineering pack
Last verified 8/28/2026
What it changes
Passed 12 of 19 test cases with the skill, and 12 of 19 without it.
Without skill61%
With skill65%
Checks it passed that the model alone missed
Includes a section for Deployment readiness.
Excludes zero-finding categories from the findings table.
Explains the deployment risk of broken external dependencies.
and 2 more in the full report
4 fixed11 passed either way5 still failing3 broke
Pass-rate lift
+4 pp
Tokens
+41%
Agent turns
0%
Response time
+9%
Model gemini-3.6-flashMeasured 8/28/2026v115 of 23 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 4
Prompt
We identified that `./public/index.html` loads font stylesheets from `http://dev-server.local/fonts.css` and an external CDN script. Which audit category does this belong to?
What was checked, without and with the skill
Checked
Without skill
With skill
Categorizes local development endpoints under external dependencies and link risk.
Missed
Passed
Explains the deployment risk of broken external dependencies.
Missed
Passed
case-16 · gemini-3.6-flash · 8/28/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md97 lines3 files
Clone Audit
Audit the requested website-clone workspace and produce an evidence-based
deployment-readiness report. Inspect the current target; never reuse findings
from an earlier project or run.
Establish scope
Confirm the target root and the intended output language.
Identify any supplied source site, screenshots, design tokens, or other
fidelity references. If none are available, mark visual fidelity as not
checked instead of guessing.
Inventory relevant HTML, CSS, JavaScript or TypeScript, assets, metadata,
configuration, and dependency manifests. Respect explicit exclusions.
Prefer static inspection. Do not execute untrusted project code, install
packages, or make network requests unless the user authorizes it.
Treat references/source-1-CLONE_AUDIT.md only as historical provenance. Do
not copy its paths, counts, or findings into a new audit unless the current
target independently confirms them.
Run the checks
Inspect each category and record the evidence used:
Fidelity assets and styles — compare against supplied references for
missing or substituted fonts and images, broken asset paths, incorrect
colors, and materially different layout or styling.
Tracking scripts and pixels — find analytics, tag managers, advertising
pixels, telemetry beacons, and unexpected third-party scripts.
Source-brand residue — find source brand names, domains, metadata,
social links, asset paths, comments, and copy that should have been replaced.
Language residue — find unintended text in languages outside the target
locale, excluding code identifiers and legitimate proper nouns.
TODOs and placeholders — find TODO or FIXME markers, lorem ipsum,
template copy, dummy links, test credentials, and unfinished states.
External dependencies and link risk — inspect remote URLs, CDNs,
localhost or development endpoints, external fonts and media, package
downloads, and dependencies that may fail, leak data, or violate deployment
constraints.