Back to skill

Security audit

rotifer-guide

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: scanners found behavior generally aligned with the skill’s Rotifer purpose, but flagged actions that may happen without the promised approval and the scan was only partially complete.

The verdict is benign, but it does not mean risk-free. Before installing, consider that the scan reported commands that may install software or submit work without the approval the skill description promises. Skillspector’s review was partial because some references could not be resolved, so the evidence is incomplete.

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
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

Installing the Rotifer CLI is relevant supporting setup, and checking its version is consistent with the declared scope. However, the supplied shell commands execute installation unconditionally rather than merely proposing it or waiting for approval as promised. No approval mechanism is present in the supplied chunk.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

The commands remain within the declared Rotifer scope, but the unconditional rotifer arena submit hello-world does not implement the promised approval boundary. The comment asserting that code and results stay local is not evidence of the command's actual data-handling behavior; that would require inspecting the CLI implementation.

Content

No source excerpt is available for this finding.

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: rotifer-guide
description: >-
  Entry point for Rotifer Protocol. Onboards a user to Rotifer Genes, scaffolds a Gene from a
  description, diagnoses a Gene's F(g) score or compile failure, searches the Rotifer Gene
  registry, upgrades a Gene's fidelity from Wrapped to Native, and submits a Gene to the Arena
  and follows its ranking. It works by proposing `rotifer` CLI commands; the ones that install
  software, send work to Rotifer's servers, publish a Gene or overwrite files wait for the
  user's approval. Meant to be invoked explicitly when working with Rotifer, or when unsure
  which Rotifer capability

MCP server referenced without pinned version: 'npx also'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx does'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.