Back to skill

Security audit

Cloudflare R2

Security checks for vulnerabilities and agentic risk

Overview

This R2 integration needs review because its safety rules permit data transfers and access grants without clear approval, and its CLI dependency can modify installed agent skills.

Before installing, review and verify the CLI installer and its automatic skill synchronization. Use narrowly scoped R2 credentials. Require explicit authorization for transit-storage copies and signed access URLs, including their target, permitted operation, recipient, and expiry. Confirm transit-storage retention and deletion terms before handling sensitive objects. The remote installer and hosted connector implementation were not verified.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied command installs software via a remotely fetched shell script rather than performing Cloudflare R2 operations. Nothing in the supplied chunk establishes that this installer is a supporting dependency for R2 functionality. The installer’s contents and subsequent behavior are not visible.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

The setup instructions pipe a remotely fetched shell script directly into bash, and the PowerShell alternative similarly evaluates downloaded code without integrity verification. Compromise of the installer source or distribution infrastructure could therefore execute attacker-controlled code with the invoking user's privileges; limiting this to first-time setup reduces exposure frequency but does not remove the risk.

Content

Scanner excerpt · SKILL.md (reported line 94)May include surrounding context.

The untagged download action uploads object data to transit storage despite the safety guidance classifying untagged actions as reads.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The download action explicitly copies R2 object contents into connector transit storage, but the safety section categorizes all untagged actions as reads that are safe to run directly. This can cause sensitive object contents to be persisted outside R2 without approval of that additional storage destination, although the third-party connector architecture is disclosed.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% 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 115)May include surrounding context.

md
- **`OOMOL_INSUFFICIENT_CREDIT`** — OOMOL billing stop. Recharge at `https://console.oomol.com/billing/token-recharge` before retrying.

- **HTTP 402 without `OOMOL_INSUFFICIENT_CREDIT`** — the third-party account itself has no active plan or credit. Report the provider error message as is; do not ask the user to recharge OOMOL.

## Resources