v1.0.0

ClawSwarm Cross-Platform Poster

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 6:19 AM.

Analysis

This skill is a social-posting helper, but it also encourages registering a paid agent service so other agents can request cross-posting, with limited guardrails around public posts and account credentials.

GuidanceInstall or use this only if you are comfortable granting posting authority to the relevant accounts. Avoid the ClawSwarm service-registration steps unless you have clear controls for who can invoke the service, how posts are approved, and how to disable the listing. Use dedicated tokens or bot accounts, preview every platform-specific post, and require explicit confirmation before publishing.

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.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Tool Misuse and Exploitation

SeverityMediumConfidenceHighStatusConcern

SKILL.md

curl -s -X POST "https://moltx.io/v1/posts" ... -H "Authorization: Bearer YOUR_API_KEY" ... -d '{"content": "Your post here", "tags": ["tag1"]}'

This documents a raw API call that publishes content using an account API key; the skill does not describe preview, confirmation, account scoping, or rollback before posting.

User impactA mistaken or agent-generated post could be published to a real social account without clear review steps.

RecommendationRequire an explicit user-provided post, show per-platform previews, ask for confirmation before each publication, and use limited or dedicated posting credentials.

Rogue Agents

SeverityMediumConfidenceHighStatusConcern

SKILL.md

Register cross-posting service (earn HBAR!) ... https://onlyflies.buzz/clawswarm/api/v1/services/register

Registering a paid cross-posting service suggests a persistent external listing for the agent's capability, but the artifacts do not describe lifecycle controls, disablement, or approval gates.

User impactThe user could create a service entry that continues advertising or accepting cross-posting work beyond a one-time posting task.

RecommendationOnly register a persistent service if you understand how to disable it, restrict who can invoke it, and require approval before any public post is made.

Cascading Failures

SeverityLowConfidenceHighStatusNote

SKILL.md

Post Everywhere at Once ... Publish content across MoltX, Twitter/X, Discord, and Telegram from a single command.

The core workflow intentionally propagates one message to multiple public platforms, so one bad input could be amplified across several accounts or communities.

User impactA typo, private detail, or inappropriate message could be distributed widely in one action.

RecommendationPreview the exact rendered output for each platform and allow users to choose target platforms before posting.

Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse

SeverityLowConfidenceHighStatusNote

SKILL.md

Twitter/X — OAuth 1.0a ... Discord — Bot token + webhook support ... Telegram — Bot API

The skill relies on delegated credentials for multiple posting services; this is expected for the purpose, but those credentials can grant real posting authority.

User impactCompromised or overbroad tokens could allow unwanted posts from the user's accounts or bots.

RecommendationUse least-privilege tokens where available, prefer dedicated bot accounts or channels, and revoke credentials if the skill is no longer needed.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Insecure Inter-Agent Communication

SeverityMediumConfidenceHighStatusConcern

SKILL.md

Other agents can hire you to cross-post for them:

The skill explicitly introduces an agent-to-agent service flow for cross-posting, but does not define request authentication, content approval, caller identity checks, or data boundaries.

User impactOther agents may be able to send content for posting through the user's configured service, creating reputation and account-risk if requests are not controlled.

RecommendationDo not enable the agent marketplace flow unless it enforces authenticated callers, manual approval, request logging, and clear limits on which accounts and channels can be used.