NotFair Google Ads Agent

ReviewAudited by ClawScan on May 14, 2026.

Overview

The skill is coherent for Google Ads auditing and management, but users should notice that it installs an external NotFair plugin and connects to live Google Ads accounts where approved changes can affect spend.

This appears to be a legitimate, purpose-aligned Google Ads assistant with explicit approval requirements for writes. Before installing, make sure you trust the NotFair plugin and hosted MCP service, connect only the intended Google Ads accounts, and carefully review any proposed budget, campaign, keyword, or deletion changes before approving them.

Findings (5)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Users are trusting an external plugin that was not included in this review package.

Why it was flagged

The skill installs a separate Node plugin, while the supplied artifact set contains only SKILL.md and no plugin source code to inspect.

Skill content
openclaw plugins install openclaw-notfair
Recommendation

Install only if you trust NotFair and the OpenClaw plugin source; review the plugin package and permissions if available.

What this means

The connected service may be able to view Google Ads account information and, after approval, perform account changes.

Why it was flagged

The setup requires authenticating and connecting Google Ads accounts, which is expected for this purpose but grants access to live advertising account data and authority.

Skill content
openclaw notfair login
openclaw notfair connect
Recommendation

Connect only the intended Google Ads accounts and use the least-privileged account access available.

What this means

Incorrect tool names or arguments could cause unintended Google Ads changes if the user approves them without review.

Why it was flagged

The generic MCP tool interface can invoke specific Google Ads tools by name and arguments; the skill mitigates this by instructing the agent to present proposed actions before writes.

Skill content
notfair_google_ads_tool - call a specific NotFair MCP tool by exact name and JSON arguments.
Recommendation

Before approving any write, check the target account, campaign, budget, keywords, and whether the change is reversible or can be created as a draft/paused change.

What this means

Google Ads account details and requested actions may be processed by a third-party hosted service.

Why it was flagged

The skill relies on a hosted MCP server to mediate Google Ads analysis and actions, so advertising account data and requests may flow through NotFair's service.

Skill content
propose approval-gated campaign changes through NotFair's hosted MCP server
Recommendation

Review NotFair's privacy, security, and Google Ads access practices before connecting sensitive advertising accounts.

What this means

Generated analysis scripts can inspect account data and produce reports, but the documented intent is read-only analysis.

Why it was flagged

The skill supports running JavaScript analysis against Google Ads data. The artifact describes this as read-only and central to the audit workflow.

Skill content
notfair_run_script - run read-only JavaScript analysis against Google Ads data.
Recommendation

Use the script tool for reporting and diagnosis only, and keep write operations behind explicit user approval.