Back to skill
Skillv1.0.0
ClawScan security
Adagent Google Ads · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 30, 2026, 6:51 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill is internally consistent: it delegates Google Ads work to an external service (AdAgent) via an MCP connector link (a credential) and requires no local installs or unrelated credentials.
- Guidance
- This skill delegates all Google Ads actions to the third-party service AdAgent and requires you to paste an MCP connector link that contains an embedded token — treat that link like a password. Before installing: (1) confirm you trust adagent.10xboost.org and review its privacy/terms, (2) ensure the Google OAuth scopes granted to AdAgent are limited to what you need (least privilege), (3) do not paste the MCP link in public chats or logs, and (4) monitor any campaign creation or budget changes the skill makes and revoke access at AdAgent if you see unexpected activity. If you prefer more control, consider using a skill that requires your own explicit API keys or OAuth flow you control.
Review Dimensions
- Purpose & Capability
- okThe name/description (manage Google Ads) matches the instructions: all operations (list accounts, create/pause/enable campaigns, keyword research, performance) are performed via the AdAgent service using an MCP connector link. No unrelated credentials or binaries are requested.
- Instruction Scope
- okSKILL.md confines runtime actions to calling the external AdAgent API (via the MCP link) and mapping user intents to API calls. It does not instruct the agent to read local files, unrelated env vars, or transmit data to unexpected endpoints. It explicitly states the MCP link is a credential and how tokens are handled on the service side.
- Install Mechanism
- okInstruction-only skill with no install spec and no code files — minimal disk footprint and low installation risk.
- Credentials
- noteNo environment variables or platform credentials are declared, but the skill requires an MCP Connector link that embeds an auth token. This credential is proportional to the stated purpose, but because it is an embedded token, users must treat it like a secret and avoid pasting it into public chat or insecure places.
- Persistence & Privilege
- okThe skill does not request always:true and does not ask to modify other skills or system-wide settings. It can be invoked autonomously by the agent (platform default), which is expected for an integration skill.