Security audit

并行研究助手免费版

Security checks for vulnerabilities and agentic risk

Overview

This skill is a file-backed research assistant that openly stores research notes locally and does not show hidden, destructive, or deceptive behavior.

Install only if you are comfortable with research topics and notes being saved under ~/.research-workspace. Avoid using callback_url unless you fully trust the destination, and manually review or delete generated files on shared or managed machines.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Behavioral ASTexec() Call, eval() Call, Dynamic Import
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill explicitly instructs the agent to create and update files under `~/.research-workspace/research/<topic-slug>/`, but it does not clearly warn users up front that invoking the skill causes persistent local filesystem writes. In an agent context with `read`, `glob`, `grep`, and especially `exec` available, silent persistence can expose sensitive research topics or leave unintended local artifacts, making this a genuine consent and privacy issue.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The documented `callback_url` allows data related to the research task to be transmitted to an external endpoint, but the skill provides no privacy, trust-boundary, or destination validation warning. In practice, this could lead users or downstream agents to send sensitive prompts, findings, metadata, or status updates to attacker-controlled infrastructure without informed consent.

Missing User Warnings

Low

Confidence: 80% confidence
Finding: The skill persists `research.md`, `prompt.md`, and optional `research.pdf` to disk, but it does not clearly state retention, visibility, or cleanup expectations. While persistence is central to the skill's purpose, lack of privacy guidance can still cause inadvertent storage of sensitive research subjects or exported documents on shared systems.

Static analysis

No suspicious patterns detected.