Openclaw Meta Ads
ReviewAudited by ClawScan on May 1, 2026.
Overview
This docs-only Meta Ads skill is coherent and safety-conscious, but users should notice it may involve Meta account tokens, ad-management permissions, and sensitive lead data.
This skill appears safe to install as a documentation-only Meta Ads helper. Before using it, make sure any Meta token is least-privilege, keep write permissions disabled unless needed, require explicit approval for every live ad change, and redact or minimize lead data before sharing or storing it.
Findings (3)
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.
A token with broad permissions could read ad account data, retrieve lead information, or make live account changes if granted management scope.
The skill documents use of Meta account credentials and permissions, including possible write permissions and lead access. This is expected for a Meta Ads skill, but it is high-impact authority.
Common permissions include: - `ads_read` for reporting and insights - `ads_management` for live changes - lead access permissions for form lead retrieval where applicable
Use least-privilege access: prefer read-only `ads_read` unless live changes are truly needed, and grant lead or management permissions only for specific approved tasks.
If an agent misinterprets approval, live campaigns or budgets could be changed in a real advertising account.
The skill contemplates live Meta Ads mutations such as pausing ads, enabling ads, editing campaigns, changing budgets, or changing creatives. It also clearly requires explicit approval first.
require explicit approval for pausing, enabling, editing, budget changes, or creative changes
Require a clear per-change confirmation that names the account, campaign/ad set/ad, field to change, old value, new value, and rollback plan before allowing writes.
Lead names, contact details, or other personal information could enter the agent conversation or project context if not minimized.
The skill may retrieve Meta lead-form data, which can contain personal information. The docs recognize this and recommend careful, minimal handling.
When lead forms are in play: - identify the relevant form IDs - retrieve leads carefully - treat lead payloads as potentially sensitive - prefer minimal necessary fields when analyzing quality or flow
Retrieve only the fields needed, redact personal details before broader analysis, and avoid saving lead payloads in shared notes or long-term memory unless explicitly approved.