Back to skill

Security audit

PDF Vector

Security checks for vulnerabilities and agentic risk

Overview

The PDF tools fit the stated purpose, but automatic setup runs unchecked remote code through a CLI that also documents persistent changes to agent skills.

Review and approve CLI installation separately before enabling this skill. Check which persistent agent skills the chosen CLI release installs or synchronizes, and disable synchronization if unwanted. Document requests pass through OOMOL to PDF Vector and may consume service credit. The bundled skill was fully inspected, but the external installer and CLI runtime were not verified; no malicious behavior was established.

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 (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied chunk performs remote script execution rather than PDF Vector searching or reading. Although installation can be a supporting implementation detail, no connection between this Oomol installer and PDF Vector is established in the supplied content. The downloaded script's contents and any further effects are unknown.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

The supplied code performs only authentication, not the declared PDF Vector data operations. Authentication could be a supporting implementation detail if tied to PDF Vector, but the supplied chunk establishes no such connection and contains no subsequent search or read behavior.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
99% confidence
Finding

Piping a remotely fetched installer directly into bash executes mutable network-delivered code without checking its contents, version, signature, or trusted digest. HTTPS protects transport but not against compromise of the legitimate distribution service, and the resulting code could access local files and credentials or establish persistence under the user's account.

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

First-time setup instructs the agent to execute remotely downloaded installation scripts without a system-integrity warning or user approval.

Medium
Category
Output Handling
Confidence
98% confidence
Finding

The missing-CLI fallback directs execution of downloaded shell or PowerShell scripts without requiring user approval, inspection, or integrity verification. Although dependency setup is relevant and conditional, it expands a document-reading task into local code execution; a compromised installer or distribution endpoint could run arbitrary commands with the invoking user's privileges.

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 78)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