Back to skill

Security audit

Spraay × Bankr

Security checks for vulnerabilities and agentic risk

Overview

This skill helps with real crypto payouts, but its behavior is disclosed and it repeatedly requires validation and user confirmation before funds move.

Install only if you intend to let an agent help prepare Bankr-funded Spraay payouts. Treat it like any real-money workflow: verify token, chain, recipient addresses, amounts, fees, and transaction details yourself before confirming execution, and keep the Bankr API key limited to Bankr use.

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
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
91% confidence
Finding

The trigger phrase "batch send" is overly generic and can match many unrelated user intents, causing this high-impact financial skill to activate outside the intended Bankr/Spraay context. Because the skill can lead to token transfers and payroll-like actions, accidental invocation increases the chance of misrouting users into a funds-moving workflow without sufficient contextual fit.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% 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 83)May include surrounding context.

md
3. Validate → estimate → execute the batch (same free-then-paid flow as
   Recipe 1). USDC or the fee token itself both work.
4. For recurring splits, re-run on a schedule the user controls; never
   auto-execute without a standing, explicit instruction.

## Recipe 3 — Treasury → Payroll

Static analysis

No suspicious patterns detected.