ClawHub
Back to skill

Security audit

Punting Buddy: Horse Racing Analysis

Security checks across malware telemetry and agentic risk

Overview

This is a read-only horse-racing analysis skill that clearly discloses its use of The Racing API and does not include executable code or live betting automation.

Install this only if you want horse-racing analysis that can fetch racecards and results from The Racing API. Store the API username and password as environment secrets, avoid pasting credentials into chat, and treat betting suggestions as informal analysis rather than guaranteed picks or placed bets.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description includes broad trigger phrases like 'what races are next', 'what is on today or tomorrow', and 'today's results', which overlap with common conversational requests. In an agent router, this can cause the skill to activate for generic scheduling or sports-result queries outside the intended horse-racing scope, leading to unnecessary data access, user confusion, or incorrect delegation.

Vague Triggers

Low
Confidence
85% confidence
Finding
The 'Use this skill when' section lists positive triggers but provides no exclusion conditions or negative examples, so a dispatcher has little guidance on when not to use it. This increases the chance of over-broad invocation on adjacent queries, especially because the skill is conversational and framed as a general 'mate' for analysis rather than narrowly bounded horse-racing tasks.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The default prompt is broad and maps generic horse-racing requests such as 'what races are next' or 'what is on today' to this skill, which can cause implicit invocation in situations the user did not clearly intend. Because the skill is read-oriented and domain-specific, the main risk is misrouting, unexpected data access, or confusing responses rather than direct system compromise, but allow_implicit_invocation makes accidental triggering more likely.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.