Back to skill

Security audit

Geokeo

Security checks for vulnerabilities and agentic risk

Overview

The geocoding workflow fits its purpose, but automatic installation and undisclosed changes to agent skills need review.

Review and approve CLI installation separately, preferably using a verified fixed version. Consider disabling its documented startup skill synchronization with OO_SKILLS_SYNC_DISABLED=1. Addresses and coordinates pass through OOMOL to Geokeo, so submit only information you intend to share. The bundled skill was fully inspected, but external installer code and CLI binaries remain unverified; no malicious intent 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
98% confidence
Finding

The visible command executes an installer hosted on cli.oomol.com, with no demonstrated connection to Geokeo functionality. Nothing in the supplied chunk establishes this as a supporting implementation detail for Geokeo access. The downloaded script's contents and any further effects are not available for assessment.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

Piping an unpinned HTTPS response directly into Bash executes whatever the installation endpoint returns, without an independent integrity or authenticity check. HTTPS protects transport but does not prevent a compromised publisher or distribution endpoint from delivering code that accesses local credentials or modifies files; this is the same underlying installation risk identified in the setup finding.

Content

Scanner excerpt · SKILL.md (reported line 56)May include surrounding context.

The geocoding workflow lacks a privacy warning about transmitting addresses and coordinates to external services.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The workflow instructs the agent to submit action payloads through OOMOL to Geokeo, including address queries or coordinates. Although the services are named, the skill does not warn that potentially sensitive location data will leave the user's environment; describing these actions as reads does not address that privacy impact.

Content

No source excerpt is available for this finding.

The setup workflow executes downloaded installation scripts without warning about their system impact.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The missing-CLI fallback directs execution of downloaded shell or PowerShell code without requiring explicit installation approval or verification of the artifact. This can introduce local system changes and execute malicious code with the invoking user's permissions if the distribution channel is compromised; the conditional setup path and declared tool restriction reduce reachability but do not make the procedure safe.

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 77)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