Security audit

Alibabacloud Network Ga Deploy Acceleration

Security checks across malware telemetry and agentic risk

Overview

The skill matches its Alibaba Cloud GA deployment purpose, but it needs Review because it can create or delete paid cloud network resources and includes under-scoped credential and cleanup guidance.

Install only if you intend to let an agent assist with Alibaba Cloud GA operations. Use a dedicated least-privilege RAM user or temporary credentials, configure secrets outside the chat, avoid passing access keys directly in commands, review every Aliyun command before execution, and explicitly verify account, resource IDs, billing impact, and deletion impact before proceeding.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
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

Findings (11)

Context-Inappropriate Capability

Medium

Confidence: 89% confidence
Finding: Requiring `aliyun sts GetCallerIdentity` and displaying AccountId, PrincipalId, and Arn exposes sensitive identity metadata that is not strictly necessary to create a GA instance. In an agent context, this increases unnecessary collection and presentation of account-identifying information, which could aid reconnaissance, logging leakage, or cross-tenant data exposure.

Context-Inappropriate Capability

Medium

Confidence: 74% confidence
Finding: The guide documents elevated and cross-account authentication modes such as RamRoleArn and RamRoleArnWithEcs without showing GA-specific justification or least-privilege boundaries. In an agent skill context, this can normalize overprivileged credentials and increase the blast radius if the skill or surrounding automation is misused or compromised.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: Using the trigger word `GA` is excessively broad and likely to match unrelated conversations, causing accidental invocation of an infrastructure-deployment skill. In an agent system with execution capabilities, unintended activation can lead to inappropriate credential checks, account queries, or cloud operations in the wrong context.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The cleanup section provides a destructive `DeleteAccelerator` command without an adjacent warning or explicit confirmation requirement. In an agent-assisted workflow, this increases the chance of accidental deletion of production networking resources and service disruption.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The guide instructs users to pass access keys directly on the command line and via environment variables without warning about exposure through shell history, process listings, terminal logs, or CI/CD job output. In automation and agent-driven environments, these channels are commonly captured and persisted, making credential leakage a realistic risk.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The document includes a resource deletion command (`DeleteAccelerator`) without any caution about irreversibility, outage risk, or the need for explicit user confirmation. In an agent skill context, this increases the chance that an automated workflow or user will execute a destructive operation without understanding the operational and billing consequences.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The file documents additional delete operations (`DeleteIpSets`) without warning that removing acceleration regions can disrupt traffic routing and invalidate assigned accelerated IPs. In infrastructure automation, omission of these cautions can lead to accidental production impact.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: `DeleteListener` is presented as a normal reference command without a warning that deleting a listener can immediately stop inbound traffic handling for associated ports and protocols. In a deployment skill, this is operationally dangerous because users may treat all listed commands as equally safe to run.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: `DeleteEndpointGroup` lacks any note that removing endpoint groups can break backend routing and failover behavior. Because endpoint groups are core traffic destinations, undocumented deletion risk can cause immediate availability problems.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The `DeleteForwardingRules` command is listed without warning that routing behavior may change instantly, potentially blackholing or misdirecting application traffic. This is especially risky in an agent skill because rule changes can be applied programmatically at scale.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The workflow demonstrates service activation and creation of billable resources, including `OpenAcceleratorService`, `CreateAccelerator`, and `--AutoPay true`, without prominently warning about account-level changes and cost implications. In an agent-driven environment, this can lead to unintended spend or unauthorized provisioning.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.