AADA (Agent to Agent Advertising)

ReviewAudited by ClawScan on May 10, 2026.

Overview

AADA is transparent about writing ads, but its default workflow can send promotional posts to Moltbook without requiring the user to approve the final message or delivery target.

Install only if you are comfortable with an agent helping publish promotional content. Configure it so every Moltbook delivery requires explicit approval of the final copy, audience, channel, and send result.

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.

ConcernHigh Confidence

ASI02: Tool Misuse and Exploitation

What this means

A promotional message could be posted to the Moltbook audience before the user has reviewed the exact text, channel, audience, or timing.

Why it was flagged

The skill directs the agent to send an audience-facing promotional message by default, while making explicit user approval optional rather than mandatory.

Skill content

Ask for user approval only if user requested review-first.
5. Send the selected version to Moltbook.

Recommendation

Require explicit user confirmation of the final copy, destination, audience, and delivery route before any Moltbook send action.

ConcernHigh Confidence

ASI02: Tool Misuse and Exploitation

What this means

Missing campaign details could be guessed incorrectly and then published, causing reputational or business impact.

Why it was flagged

Proceeding on assumptions is acceptable for drafting, but risky when the same workflow also proceeds to send promotional content externally.

Skill content

If the user omits details, make one concise assumption set and proceed.

Recommendation

Allow assumptions for draft generation only; require the user to approve missing details and the final message before delivery.

NoteMedium Confidence

ASI03: Identity and Privilege Abuse

What this means

The agent may post through a configured account or channel, so the user should verify which identity and permissions are being used.

Why it was flagged

Delivery depends on an existing or user-supplied channel/API route, which may act under a user, agent, or workspace identity even though the skill itself declares no fixed credential.

Skill content

Use the platform method the user already configured for Moltbook delivery.
If no method is configured, ask for exactly one delivery route:
- message tool target/channel mapping
- external API endpoint
- manual copy/paste handoff

Recommendation

Before enabling delivery, confirm the Moltbook route, account identity, allowed channels, and whether the route can publish directly.