Security audit

项目管理API

Security checks across malware telemetry and agentic risk

Overview

This skill is a project-management GraphQL helper, but it asks for broad command/file authority and can mutate external work data without clear scoping or confirmation guidance.

Review this before installing. Use it only with a least-privilege project-management API key and only for accounts where automated issue, project, label, cycle, view, and comment changes are acceptable. Avoid letting it run from vague prompts, and require explicit confirmation for mutations, deletes, archives, or command execution.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Context-Inappropriate Capability

Medium

Confidence: 94% confidence
Finding: The skill declares `exec` alongside `read` and `write` even though the documented purpose is a project-management GraphQL integration. That mismatch expands the attack surface significantly: if an agent uses this skill on untrusted input, generic command execution could be abused to run arbitrary shell commands, access local secrets, or modify files unrelated to the stated API task.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The invocation description is broad and ambiguous, describing generalized project-management processing rather than tightly scoped operations. In an agent setting, vague activation criteria can cause the skill to be selected in contexts the user did not intend, increasing the chance of unnecessary access to external systems or triggering side-effecting operations.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill exposes `write` and `exec` capabilities but does not prominently warn users that it may alter local files or external project data. In agent workflows, missing warnings reduce informed consent and make it easier for destructive or unintended actions to occur when the skill is invoked automatically or through ambiguous prompts.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.