ClawHub
Back to skill

Security audit

Long Run Hybrid Coach

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Garmin/Intervals training coach integration that reads fitness evidence and updates workouts only through previewed, user-confirmed changes.

Install only if you are comfortable connecting the coach to Garmin/Intervals or a similar training provider and letting it read training history, weight entries, and planned workouts. Review each preview carefully before confirming calendar or workout changes.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description says to use the skill for a wide range of common requests such as '月目標', '周計畫', '今天練什麼', and '記錄體重', which are short, everyday phrases that could overlap with ordinary conversation. Although some examples are domain-specific, the file does not define explicit trigger constraints or negative examples beyond two exclusions, making activation scope ambiguous.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
91% confidence
Finding

The instruction to 'retry unchanged approved effects without asking again' permits repeated write-side actions based on prior approval, even after delivery failures, state changes, or user intent changes. In a skill that can modify training plans and push calendar/workout effects to external systems, this weakens per-action consent and can cause unintended schedule changes or repeated delivery attempts without fresh authorization.

Content

Scanner excerpt · SKILL.md (reported line 101)May include surrounding context.

md
intermediate JSON.
8. Show every calendar effect in an exact preview and obtain one explicit
   confirmation. A plan change may include delivery and withdrawal in that same
   preview; retry unchanged approved effects without asking again. Report only
   delivery the product observed — Intervals accepting a workout is never
   evidence that Garmin Connect or the watch received it, and a strength day
   reaches the calendar as a title carrying no executable structure.