Back to skill

Security audit

ReelsFarm Social Scheduler

Security checks for vulnerabilities and agentic risk

Overview

The skill supports approved social-media publishing with clear account limits, previews, and user confirmation.

Protect the API key and approve the exact account, content, caption, privacy settings, and posting time before execution. Cancellation should also require your approval. The complete Markdown artifact was inspected; the separately downloaded CLI and live service were not audited.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

Listing posts and checking status could support a publishing workflow. However, the supplied chunk only manages existing posts and includes cancellation, which can prevent publication and is not declared. It contains no scheduling or publishing commands.

Content

No source excerpt is available for this finding.