Security audit

Solo Workflow Engine

Security checks for vulnerabilities and agentic risk

Overview

This is mostly a no-code automation guide, but it asks for broad command and file permissions that are not well scoped for that purpose.

Review this skill before installing if your agent environment grants tools from the skill metadata. It is suitable as guidance for planning no-code automations, but you should avoid allowing unrestricted command execution or file writes unless you explicitly need them, and approve any workflow that changes CRM records, sends posts, creates invoices, or touches payment data.

A.I.G

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

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Behavioral ASTexec() Call, eval() Call, Dynamic Import
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

Findings (5)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill declares both write and exec capabilities despite being presented mainly as a workflow methodology and guidance document, and it does not clearly warn users that the agent may affect the local system. Overprivileged capability declarations increase the blast radius if the skill is misused, modified, or combined with prompt injection, because the agent can move from advice into file-system or command execution.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The skill tells the agent to run a shell-level network diagnostic command (ping) when required parameters are missing, which is unrelated to the stated purpose of providing workflow methodology and configuration guidance. In an agent environment with exec enabled, this expands the skill from documentation/advice into host interaction, creating unnecessary command-execution behavior and potential abuse paths for environmental probing.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The troubleshooting guidance instructs shell command execution without warning about local system impact, network access, or the fact that such commands will run on the agent host. Even a seemingly harmless command like ping normalizes command execution from markdown instructions and can be adapted into broader unsafe operational behavior in a skill that already declares exec.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The summary and description, along with the rest of the skill instructions, are presented exclusively in Chinese, effectively forcing a specific language for users. The file does not provide a language/locale choice or clearly justify that this is a region-specific skill with an explicit locale constraint.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 77% confidence
Finding: Lines L622-L624 assert that the skill is purely Markdown-based and does not involve API key handling, but other sections describe operational behaviors such as callback URLs, API-driven interactions, and command execution support. While not proof of credential access in code, the documentation presents conflicting intent about whether this is only static guidance or an operationally active skill.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.