ClawHub
Back to skill

Security audit

Glovo

Security checks for vulnerabilities and agentic risk

Overview

This Glovo skill is purpose-aligned and includes clear safeguards for browser control, cart changes, checkout, and local memory, though users should be careful about opting into persistent notes.

Install only if you want an agent to help with real Glovo browsing and ordering. Before using it, decide whether it may control a signed-in browser session and whether it may keep local notes in ~/glovo/ about addresses, stores, and ordering habits; do not allow it to save passwords, payment card details, or verification codes.

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)

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 117)May include surrounding context.

md
- Treating the home page as actionable before an address is set -> store availability and fees are unreliable.
- Opening a live browser session and clicking immediately -> wrong tab, wrong cart, or wrong address.
- Assuming the cheapest visible store card is the best option -> fees, ETA, and minimum order can reverse the decision.
- Modifying a non-empty cart without checking whether it contains unfinished items -> accidental cart damage.
- Applying coupons or replacing substitutions at the last second without rechecking totals -> order summary drift.
- Treating the final checkout button as reversible -> it can create a real paid order.

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 161)May include surrounding context.

md
- Treating the home page as actionable before an address is set -> store availability and fees are unreliable.
- Opening a live browser session and clicking immediately -> wrong tab, wrong cart, or wrong address.
- Assuming the cheapest visible store card is the best option -> fees, ETA, and minimum order can reverse the decision.
- Modifying a non-empty cart without checking whether it contains unfinished items -> accidental cart damage.
- Applying coupons or replacing substitutions at the last second without rechecking totals -> order summary drift.
- Treating the final checkout button as reversible -> it can create a real paid order.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The setup directs the agent to determine activation behavior 'within the first few exchanges' using broad triggers like 'whenever Glovo comes up' or 'for real browsing and ordering work.' That ambiguity can cause the skill to activate in situations the user did not clearly intend, increasing the chance of inspecting live sessions, handling carts, or progressing toward checkout without sufficiently explicit consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs the agent to record activation rules in ~/glovo/memory.md, but it does not require a user-facing disclosure or consent before persisting preferences. Because the stored data includes behavioral preferences, addresses, neighborhoods, and ordering habits, this creates a privacy risk and can lead to retention of sensitive personal context without the user's informed approval.

Content

No source excerpt is available for this finding.