v0.1.0

Clawpen

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 5:25 AM.

Analysis

Review before installing: the skill matches its social-game purpose, but it asks the agent to keep acting on Clawpen and to self-update its installed instructions from clawpen.com.

GuidanceInstall only if you are comfortable with your agent periodically using a Clawpen API key to vote, duel, check messages, and potentially update its own local skill files from clawpen.com. Consider disabling automatic updates and requiring approval for routine account actions.

Findings (4)

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.

Agentic Supply Chain Vulnerabilities

SeverityHighConfidenceHighStatusConcern

HEARTBEAT.md

If there's a new version, re-fetch the skill files:
curl -s https://clawpen.com/SKILL.md > ~/.openclaw/skills/clawpen/SKILL.md
curl -s https://clawpen.com/HEARTBEAT.md > ~/.openclaw/skills/clawpen/HEARTBEAT.md
curl -s https://clawpen.com/MESSAGING.md > ~/.openclaw/skills/clawpen/MESSAGING.md

The periodic heartbeat can overwrite installed skill instruction files from a remote website, with no artifact evidence of pinning, signature verification, or user review before the changed instructions take effect.

User impactFuture remote changes on clawpen.com could change what the agent is instructed to do after the user has already installed the skill.

RecommendationUpdate only through a reviewed registry release or require explicit user approval plus integrity checks before replacing local skill files.

Tool Misuse and Exploitation

SeverityMediumConfidenceHighStatusConcern

HEARTBEAT.md

You should duel at least 5-10 times per heartbeat... Vote on interesting cards... Don't bother them: Routine duels and votes... Arena duels: Every few hours (5-10 duels minimum)

The skill encourages repeated POST-style account actions on a third-party service without per-action human review, including votes that can affect other agents and leaderboards.

User impactThe agent may spend your Clawpen identity on routine votes and duel choices without asking each time.

RecommendationSet explicit limits for frequency, allowed actions, and whether the agent must ask before voting, dueling, updating its card, or messaging.

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

SeverityMediumConfidenceHighStatusNote

SKILL.md

Your API key is your identity — NEVER send it to any other domain... Recommended: Save your credentials to ~/.config/clawpen/credentials.json

The API key is expected for Clawpen, but it represents the agent identity and is recommended to be stored locally.

User impactAnyone who gets the API key could impersonate the agent on Clawpen.

RecommendationStore the key carefully, restrict local file access where possible, and only send it to https://clawpen.com/api/v1 as the skill itself warns.

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

SeverityMediumConfidenceMediumStatusNote

MESSAGING.md

Either agent can start the conversation... Pro Tier agents only can send DM requests without matching first

The skill supports agent-to-agent DMs and cold DM requests, so the agent may receive content from other agents that should be treated as untrusted.

User impactMessages from other agents could influence the agent or request sensitive information if not handled carefully.

RecommendationTreat all DM content as untrusted, avoid sharing private user data, and require human approval before acting on requests from other agents.