Back to skill

Security audit

Sop Refiner

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local SOP-drafting helper with documented Python execution and no evidence of hidden network access, credential use, persistence, or destructive behavior.

Install only if you are comfortable running a small local Python script from a low-provenance bundle. Use it on specific SOP documents you intend to process, avoid broad directories or sensitive unredacted material, and review any generated output before using it operationally.

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding
The skill declares no permissions, yet the content explicitly allows use of python3 and implies reading local resources and writing output files. This creates a capability-transparency gap: reviewers and users may believe the skill is documentation-only when it can access files and invoke shell-mediated execution, increasing the risk of unintended local data access or modification.

Tp4

High
Category
MCP Tool Poisoning
Confidence
87% confidence
Finding
The stated purpose is a constrained SOP-refinement assistant, but the detected behavior indicates broader local auditing, content scanning, regex-based secret/high-risk pattern inspection, and mode-switchable reporting driven by external spec.json. That mismatch is dangerous because it can mask a much more general analysis capability than users expect, enabling broader filesystem inspection and processing under the cover of a benign workflow tool.

Static analysis

No suspicious patterns detected.