ClawHub
Back to skill

Security audit

Maps

Security checks for vulnerabilities and agentic risk

Overview

This maps skill is a disclosed planning and provider-selection aid that may use local notes and send location data to chosen map providers, with no hidden execution or destructive behavior found.

Install only if you are comfortable with map tasks potentially using addresses, coordinates, route endpoints, and place queries. Treat `~/maps/` as sensitive because recurring places can reveal personal routines, and choose explicit activation plus confirmation-before-live-calls if you want tighter control.

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 (4)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 136)May include surrounding context.

md
- modify its own `SKILL.md`

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 153)May include surrounding context.

md
This skill NEVER:
- invent place data, ETAs, or coverage claims
- scrape undeclared providers behind anti-bot flows
- share map links externally without approval
- persist sensitive location history without telling the user first

## Related Skills

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The setup explicitly offers an 'always on' activation mode for broad categories like place, route, geocode, and map-link tasks, which weakens trigger boundaries for when the skill should engage. In a maps skill, that can lead to unintended activation on routine conversations containing location references, increasing the chance of unnecessary processing, retention, or disclosure of sensitive location-related context.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The file directs the system to establish locale defaults including a "preferred place-language bias," which can force language or locale behavior. Because the text does not require explicit user opt-in for that bias or justify a fixed locale constraint, it risks violating language/locale policy expectations.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.