Back to skill

Security audit

Anthropic

Security checks for vulnerabilities and agentic risk

Overview

This Anthropic integration needs review because its setup directs the agent to execute unverified remote installers without a clear approval step.

Review and approve a verified CLI installation separately before enabling this skill, including any bundled skills it installs. Requests pass through OOMOL, so consider the content you submit and the connected account you authorize. Retain the required message-payload confirmations. The remote installers and server-side credential handling remain unverified; the artifact does not establish malicious intent.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied code installs software through an Oomol-hosted remote script and shows no Anthropic-related operations. Nothing in the supplied content establishes this installation as a supporting implementation detail for the declared Anthropic functionality. The remote script's contents and any further effects cannot be determined from this chunk.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

The command pipes a mutable HTTPS response directly into Bash without pinning or independently verifying the executable content. Compromise of the distribution endpoint or its publishing infrastructure could yield arbitrary code execution with the invoking process's privileges; HTTPS protects transport but does not establish that the script itself is trustworthy.

Content

Scanner excerpt · SKILL.md (reported line 58)May include surrounding context.

The setup workflow executes downloaded installation scripts without warning about remote code execution or local system changes.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The setup instructions direct execution of downloaded shell and PowerShell scripts without an explicit approval step or explanation of local system changes. A missing-CLI error can therefore lead an agent from an Anthropic task into executing remote code with its local privileges; the conditional setup trigger limits exposure but does not establish authorization.

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 79)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