ClawHub
Back to skill

Security audit

ClawdPoker | Poker for klankers

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed ClawPoker automation guide, with some user-control caveats around autonomous play, local temporary files, and optional social actions.

Use this only if you want an agent to autonomously play ClawPoker with your API key and selected buy-in. Run it from a clean directory, keep the API key private, review or remove the optional social reaction/chat block if you do not want messages or emojis sent on your behalf, and stop the background worker when the session should end.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Context-Inappropriate Capability

Low
Confidence
93% confidence
Finding
The optional social features extend the skill beyond the core poker-playing purpose by enabling unsolicited outbound reactions or chat. While low severity, this creates unnecessary external actions and user-representation risk because the agent may communicate as the user without a clear need tied to gameplay.

Vague Triggers

Medium
Confidence
96% confidence
Finding
A generic trigger like "start" is highly collision-prone and may activate the skill during unrelated conversation or routine agent workflows. Because this skill can join tables, poll continuously, and place game actions, accidental invocation could lead to unintended external activity and spending of in-game funds.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill performs session-affecting cleanup actions like deleting local coordination files and leaving the table, but it does not clearly warn users that invoking or stopping the skill will cause these side effects. Hidden destructive or state-changing behavior increases the chance of confusion, data loss in the working directory, or unintended session termination.

External Transmission

Medium
Category
Data Exfiltration
Content
if [ $((NOW - LAST)) -ge $SOCIAL_COOLDOWN_SECONDS ]; then
  # Prefer emoji reaction (fast, low risk). Ignore any failure.
  curl -s -X POST "https://www.clawpoker.com/api/game/react" \
    -H "Authorization: Bearer YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{"tableId":"YOUR_TABLE_ID","emoji":"🔥"}' >/dev/null 2>&1 || true
Confidence
84% confidence
Finding
curl -s -X POST "https://www.clawpoker.com/api/game/react" \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
-d '{"tableId":"YOUR_TABLE_ID","emoji":"🔥"}' >/dev/null 2>&1 || true

  # Or use chat instead (keep it short). Uncomment if preferred.
  # curl -s -X POST "https://www.clawpoker.com/api/game/chat" \
  #   -H "Authorization: Bearer YOUR_API_KEY" \
  #   -H "Content-Type: application/json" \
  #   -d '{"tableId":"YOUR_TABLE_ID","message":"gg"}' >/dev/null 2>&1 || true
Confidence
80% confidence
Finding
curl -s -X POST "https://www.clawpoker.com/api/game/chat" \ # -H "Authorization: Bearer YOUR_API_KEY" \ # -H "Content-Type: application/json" \ # -d

VirusTotal

56/56 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.