Compl

Reviews ad content for compliance by identifying misleading claims, prohibited language, and improper imagery to ensure adherence to advertising standards.

Security Scan

This instruction-only skill is internally consistent with its stated purpose (ad compliance review) and does not request credentials, install software, or perform unexpected actions.

Purpose & Capability

The name/description (ad compliance) matches the SKILL.md, examples.md, and reference.md which contain policy guidelines and concrete bad/good ad examples. There are no unrelated requirements (no env vars, binaries, or config paths) that would be unnecessary for this purpose.

Instruction Scope

Runtime instructions are prose examples and policy references; they do not instruct the agent to read files, access system state, or send data to external endpoints. The guidance stays focused on evaluating ad text and images against policy rules.

Install Mechanism

No install spec and no code files — nothing is downloaded or written to disk. This is the lowest-risk form (instruction-only).

Credentials

The skill requires no environment variables, credentials, or config paths. That is proportionate for a purely instructional ad-compliance checker.

Persistence & Privilege

always is false and the skill does not request elevated persistence or modify other skills/config. Autonomous invocation is allowed by default but is not combined with other red flags here.

Assessment

This skill is an instruction-only guide for flagging non-compliant ad content and presents no obvious technical risks: it doesn't ask for credentials or install code. Before relying on it in production, verify the policy reference is up to date for your target ad platform/region (policies change frequently), test the guidance on representative ads, and do not treat its outputs as legally binding — consider human review or integration with official platform APIs (which would require appropriate credentials) before automated enforcement.

Ad Compliance Examples

Example 1: General Product Ad

❌ BAD (Non-Compliant):

• Headline: Get a Free Smartphone Instantly!
• Description: Hurry, only 17 remain! Click here to claim your $0 down phone today.
• Image: A phone with a big red "Download" button overlay. Reason: Uses "Free", "Instantly", fake quantities ("only 17 remain"), clickbait ("Hurry", "today"), and a fake button overlay.

✅ GOOD (Compliant):

• Headline: Compare Smartphone Features
• Description: Learn about the latest smartphone models and their specifications to find the right fit for your needs.
• Image: A clean, high-quality image of a smartphone on a desk. No text or button overlays.

Example 2: Automotive Ad

❌ BAD (Non-Compliant):

• Headline: Buy a Brand New SUV for $1,000
• Description: Search our inventory for unbelievable deals on new SUVs. Apply now!
• Image: An AI-generated futuristic concept car with a fake play button. Reason: Unreasonably cheap offer ($1000 for new SUV), uses "Search" and "Apply now" (implied functions), uses "unbelievable", and features a concept car with a fake button.

✅ GOOD (Compliant):

• Headline: 2024 Family SUVs Overview
• Description: Read our comprehensive guide on the features and safety ratings of this year's top family SUVs.
• Image: A realistic, actual photo of a standard 2024 SUV model parked on a street.

Example 3: Loan Ad

❌ BAD (Non-Compliant):

• Headline: Guaranteed $5,000 Loan Approval
• Description: Get cash in 5 minutes with no credit check. Start now!
• Image: A hand pointing directly at a stack of cash. Reason: Uses "Guaranteed", fixed amount without range, "Get cash in 5 minutes", "no credit check", "Start now", and a pointing hand image.

✅ GOOD (Compliant):

• Headline: Personal Loan Options ($1,000–$10,000)
• Description: Explore various personal loan options. Terms and availability depend on your credit profile.
• Image: A professional image of a person reviewing financial documents.