Back to skill

Security audit

Alchemy

Security checks for vulnerabilities and agentic risk

Overview

The Alchemy queries are appropriately scoped, but setup directs the agent to execute an unverified remote installer without a user approval step.

Review the CLI setup before enabling this skill. Prefer a separately reviewed, version-pinned installation and require approval before running either remote installer. Consider the CLI's broader agent-skill installation behavior and OOMOL's handling of connected credentials. The bundled artifact supports read-only Alchemy use; the remote installer's implementation could not be 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
  • 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 command installs remote software rather than searching or reading Alchemy data. Nothing in the supplied chunk establishes this installer as a supporting dependency for the declared functionality. The downloaded script's contents are unavailable, so its further effects cannot be determined.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
99% confidence
Finding

The command pipes mutable network content directly into Bash without version pinning or independent integrity verification; the PowerShell alternative uses the same download-and-execute pattern. If the distribution endpoint or release pipeline is compromised, attacker-controlled code could execute with the agent's local privileges, potentially accessing files and credentials; HTTPS alone does not prevent this supply-chain scenario.

Content

Scanner excerpt · SKILL.md (reported line 59)May include surrounding context.

First-time setup instructs the agent to execute remote installation scripts without warning about local system changes or remote-code execution.

Medium
Category
Output Handling
Confidence
96% confidence
Finding

When the CLI is missing, the skill directs the agent to install software by executing remote scripts without requiring explicit approval or explaining local system changes. This crosses the boundary from reading Alchemy data to modifying the host; the missing-command condition limits exposure but does not establish user 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 80)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