AdaptlyPost

AdvisoryAudited by Static analysis on May 3, 2026.

Overview

No suspicious patterns detected.

Findings (0)

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

If approved, the agent can publish or schedule content on connected social accounts, and mistakes may be visible publicly.

Why it was flagged

The skill can create public social media posts and upload media through API calls. This is central to the stated purpose, and the artifact explicitly warns that these are high-impact actions.

Skill content
Posts are **public, attributable, and hard to fully retract**. Treat every `POST /social-posts` and `POST /upload-urls` as a high-impact action.
Recommendation

Use the skill only with deliberate per-post approval, review the exact text, platforms, account names, timing, visibility settings, and prefer drafts for sensitive or first-time use.

What this means

A token with access to many accounts could allow broad posting or account listing through AdaptlyPost if misused.

Why it was flagged

The API key grants delegated access to connected social accounts. The artifact mitigates this by recommending a dedicated revocable token and limiting connected accounts.

Skill content
generate a **dedicated, revocable** API token for this agent... Connect only the social accounts the agent actually needs. The token has delegated access to every account in the group
Recommendation

Create a separate AdaptlyPost API token for this agent, connect only the needed accounts, and revoke or rotate the token when no longer needed.

NoteHigh Confidence
ASI08: Cascading Failures
What this means

A wrong account, typo, or visibility setting could be repeated across multiple scheduled posts if bulk actions are approved carelessly.

Why it was flagged

Bulk scheduling can amplify a mistake across many queued posts. The artifact recognizes this risk and instructs the agent to obtain explicit batch consent and start with a small batch.

Skill content
Never batch without explicit batch consent. If the user asks to schedule many posts in a row, ask them to confirm a **small first batch** (e.g. 1–3 posts) before scheduling the rest.
Recommendation

Approve small batches first, verify results in AdaptlyPost, and only then authorize larger scheduling runs.