site-audit

site-audit is a command for coding agents from SaigonXIII/evc. It costs 35 tokens per session (1,117 once invoked), scanned A, original, MIT.

A command that checks a live website's wording against a brand rule file, including banned phrases, outdated partners, required product language, and positioning.

In plain words
What is it for?
Use it to audit website text and receive a prioritised list of issues with node IDs for direct fixes in a connected site builder or CMS.
Why use it?
It finds copy that no longer matches the approved brand message before someone has to inspect every page manually.

Command

Install

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.

agentmods
npx agentmods add commands/saigonxiii/evc/site-audit
Clone the repo
git clone --depth 1 https://github.com/SaigonXIII/evc

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.

agentmods badge for site-audit

README.md
[![agentmods](https://agentmods.dev/badge/commands/saigonxiii/evc/site-audit.svg)](https://agentmods.dev/commands/saigonxiii/evc/site-audit)
Your own site
<a href="https://agentmods.dev/commands/saigonxiii/evc/site-audit"><img src="https://agentmods.dev/badge/commands/saigonxiii/evc/site-audit.svg" alt="Measured on agentmods" height="20"></a>
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,117 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00035 $0.01117
Opus 5 $0.00017 $0.00558
Sonnet 5 $0.00007 $0.00223
Haiku 4.5 $0.00003 $0.00112

Measured 5d ago against content hash 1c030416c1b4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

site-audit 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 5d 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.

_templates/commands/site-audit.md · 130 lines

How it starts

The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Site Audit

Scan the live {{WEBSITE_URL}} site for copy that conflicts with the brand canon. Returns a prioritised list of issues with the exact node IDs needed to fix them.

Usage

/site-audit

No arguments needed. Connects to the active site builder or CMS via MCP if one is available.

Prerequisites

  • A site source: either a site-builder MCP plugin showing "online", a CMS with an accessible API, or a list of public URLs you can provide
  • brand/canon-rules.json must exist and be populated (run /setup-canon first if not)

Steps

1. Load the brand canon

Read brand/canon-rules.json and hold in memory:

  • banned_phrases[] — with reasons
  • framing_red_flags[]
  • defunct_partners[] — must never appear anywhere on the site
  • live_partners[] — names the site SHOULD be consistent about
  • required_language.product_name — spelling and casing
  • required_language.company_anchor — the positioning statement that should appear on the homepage / hero
  • usps[] — the selling points the site should reinforce
  • audience + positioning — the frame the site should be written for
  • compliance_rules[<jurisdiction>] — if COMPLIANCE_JURISDICTIONS is set in config.env

If canon fields are empty, note the gap and continue with what's available. Do not audit against rules the user hasn't defined.

2. Collect site content

Source the text from whichever surface is available:

  • Site-builder MCP (Framer, Webflow, Figma Sites, etc.): list all public pages and pull text nodes with their node IDs
  • CMS API: query all published collections/items
  • URL list: if no MCP or CMS is wired, ask the user for the list of public URLs to audit and fetch them via WebFetch

For each piece of text captured, record:

  • Source ID (node ID, CMS item ID, or URL + selector)
  • Text content
  • Page/location it lives on

3. Run the audit

Apply the same checks as /content-review, at site scale.

Critical — automatic fails
  • Banned phrases: any match against banned_phrases[] is a critical fail
  • Defunct references: any appearance of a name in defunct_partners[] is a critical fail (these should be zero)
  • Compliance violations: any banned_claims for an active jurisdiction

Read the full file on GitHub · 130 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.

  1. 5d ago First seen · 130 lines · 35 tokens per session scan A 1c030416c1b4

Subscribe to this mod's changes

site-audit is a command published in the GitHub repository SaigonXIII/evc (56 stars, last pushed 2mo ago), licensed MIT. It adds 35 tokens to every session and 1,117 once invoked, about $0.0002 per session on Opus 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 2026-08-30.