Back to skill

Security audit

Arcmira

Security checks for vulnerabilities and agentic risk

Overview

The Arcmira workflow is coherent, but its setup fallback directs the agent to execute unverified downloaded code without first asking you.

Review the setup behavior before installing. Install a verified CLI release yourself and require approval for subsequent installation or agent-skill changes. Arcmira requests pass through OOMOL, and transcript retrieval can consume credits. The remote installer was unavailable for inspection, so its exact local effects remain uncertain.

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
98% confidence
Finding

The supplied code performs remote script execution rather than any visible Arcmira data operation. Nothing supplied establishes the Oomol installer as a supporting dependency for Arcmira, so this is an undeclared capability involving an inconsistent resource. The downloaded script's contents and subsequent effects are not available for assessment.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
97% confidence
Finding

Piping an HTTPS response directly into bash executes mutable remote code without integrity verification or review; the PowerShell alternative has the same property. Compromise of the installation endpoint or its delivery infrastructure could execute attacker-controlled commands with the agent's local privileges, exposing accessible files and credentials.

Content

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

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

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The missing-CLI fallback directs the agent to execute downloaded installation code without explaining its system effects or obtaining explicit installation approval. The conditional setup path limits exposure, and the declared tool allowlist may restrict execution, but neither makes the installation instructions safe if executed with shell access.

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