How does this skill differ from simply asking an LLM to check my brand guidelines?

Unlike standard prompts that rely on model 'vibes,' this skill uses the ai·rete·rag MCP server to convert your brand manual into a YAML-based policy engine, providing verifiable PASS/FAIL verdicts backed by specific manual citations.

Which AI agents or environments are compatible with this compliance checker?

This skill is designed for browser-enabled environments like Claude (via Chrome) or any agent with access to a Playwright-based MCP server to capture rendered CSS and layout data.

What specific visual and text elements can the skill audit?

The skill analyzes rendered hex colors, font-family declarations, mobile responsiveness, and copy-specific constraints like word counts and banned phrases.

How difficult is the initial setup for a new brand manual?

Setup involves calling the 'get_parameters' tool to establish your brand domain and then running 'import_policy_rules' to ingest your PDF or web-based guidelines into the logic engine.

What exactly is included in the purchase of this skill?

Your purchase includes the full logic for the brand compliance workflow, including the automated rule extraction and deterministic reporting modules, though it requires a separate ai·rete·rag API key.

Are there any limitations to what the skill can detect on a landing page?

While the skill is excellent at auditing CSS and text-based rules, it cannot automate subjective 'taste' decisions or audit colors embedded directly within flattened image files.

Brand Compliance Check

Example session with this skill installed

Check if the landing page at https://acme-testing.com/promo complies with the 'acme-inc' brand guidelines. Use the browser tool to check colors and fonts.

Read your context and instructions
Compiled the brand compliance check
Generated the UI component

I have performed a brand compliance check on the rendered page at https://acme-testing.com/promo (1440px).

Verdict: REJECT

Compliance Report

Status	Rule Name	Finding	Guideline Citation
FAIL	Approved Palette	Found #ff2d55 on hero button	"Use only the approved palette: #000000, #FFFFFF, and #00FF00."
FAIL	Font Consistency	Found 'Open Sans' rendering	"Primary brand font must be 'Inter'. Use 'Arial' only as a fallback."
PASS	Headline Length	8 words detected	"Headlines should stay under 10 words."
REVIEW	Logo Placement	Logo found in footer	"The primary logo should appear in the top-left navigation header."

Audit Details

Domain: acme-inc
Decision ID: dec_8829103af2
Source: rendered https://acme-testing.com/promo @1440px
Missing Facts: None.

Next steps

Update the primary button color from #ff2d55 to the approved brand green #00FF00.
Fix the font-face declaration in global.css to correctly load 'Inter' instead of 'Open Sans'.
Move the header logo component into the navigation bar to resolve the placement warning.

global.css

TSX · React component

Generated

Example file from a real run - the skill writes it into your workspace.

Connects securely to your tools. The creator never sees your data.

About this skill

**Turn your brand guidelines into rules, then check anything against them. ** Paste your guidelines or give the agent the page, and it drafts rules like “approved palette only”, “logo present” and “no exclamation marks in headlines”. Each rule cites the guideline sentence it comes from, and you review every one before it’s saved.

Then point it at a landing page, ad, screenshot or draft. You get a verdict per guideline (pass, fail or review) with the citation and what was found:

FAIL · CTA not Acme Blue · Found #ff2d55 · “Calls to action must use Acme Blue.”

Checks what visitors actually see. With a browser-enabled agent, it renders the page at desktop and mobile width, measures the visible colours and fonts, catches fonts that fail to load, and checks the logo from a screenshot. Without a browser, it reads the page’s code, or works from text and screenshots you share. Your agent reads the pages itself, so pages behind a login work too.

Same input, same verdict. Rules decide, not the model’s opinion, and every decision keeps an audit trail of where each fact came from.

Setup: about 5 minutes. Needs a free API key from ai-rete-rag.com (free plan: 1 brand, 1,000 checks a month). Works with Claude Code, Cursor, Claude Desktop and other MCP clients.

How to use

Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

Reviews

No reviews yet

Be one of the first to try it. Every listed skill passes our trust checks below.

Security scanned

Passed our 8-point scan before listing

Fresh listing

Recently published to Agensi

Free forever

No account required to browse

Brand Compliance Check