huawei-cloud-swr-image-governance

• Outcome: Pass
• Latest audit: Aug 22, 2026 · 12:28 PM UTC
• Version: 1.0.1

Overview

This skill is a disclosed Huawei Cloud SWR governance guide with high-impact cloud operations, but it consistently frames them as user-directed and confirmation-gated.

Install only if you intend the agent to help administer Huawei Cloud SWR. Use least-privilege IAM credentials, prefer temporary credentials, verify hcloud installer downloads before running them, and require explicit review before permission grants, retention rules, shared domains, or agency creation.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (10)

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The guide instructs users to download and immediately execute a remote shell script via curl piped into bash semantics, including a non-interactive mode, without any integrity verification, signature check, pinning, or warning about arbitrary code execution on the host. If the hosting location, DNS, TLS trust chain, or script content is compromised, users could run attacker-controlled code with the permissions of the invoking user, potentially leading to full workstation or CI runner compromise.

Missing User Warnings

Low

• Confidence: 82% confidence
• Finding: These Linux installation steps place a downloaded binary into /usr/local/bin using sudo, which modifies the system globally and elevates trust in an internet-fetched artifact without any checksum or signature verification. While common in admin documentation, it is still dangerous because a tampered binary or mistaken execution context could result in privileged persistence or execution of malicious code across the host.

Missing User Warnings

Low

• Confidence: 82% confidence
• Finding: The macOS instructions similarly direct users to install an internet-downloaded binary system-wide with sudo, but do not warn about host modification or verify the archive's authenticity. In a compromised distribution scenario, this could lead to execution and persistence of untrusted code with elevated trust on the endpoint.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The document provides copy-pastable permission-granting commands, including full-control auth values, without an explicit warning that these commands modify access control and may grant broad or namespace-wide privileges. In this skill’s context, that omission is security-relevant because users are performing IAM-like delegation on container repositories, where accidental over-granting can expose or allow modification of images across multiple repositories.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The instructions tell users to create an SWR agency delegation if one is not present, but do not clearly warn that this changes cloud account configuration and establishes a trust relationship that can enable additional service actions. In a cloud-governance skill, silently normalizing delegation creation increases the chance of unintended privilege expansion or persistent cross-service access being enabled without adequate review.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The guide includes commands to create shared download domains that grant another IAM domain read access to a repository, but it does not clearly warn that this exposes repository contents outside the current tenant/account boundary. In an image-governance skill, omission of that boundary-crossing warning can lead users to unintentionally share proprietary or sensitive container images with external parties.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The retention-rule section documents create/update/delete operations but does not prominently warn that retention policies can automatically delete tags/images once executed. In a repository governance context, this omission is dangerous because users may apply broad selectors or incorrect rules and cause irreversible loss of deployable artifacts, rollback points, or forensic evidence.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill instructs users to run `CreateAgency`, which changes account configuration by creating a delegation that can grant SWR access to other Huawei Cloud services, but it does not explicitly warn that this is a privileged, state-changing action with security implications. In an agent setting, this omission is risky because a user asking for diagnostics or setup help could be guided into granting broader access than they realize.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The documentation explicitly recommends using `--deadline=forever` for cross-organization image sharing but does not pair that guidance with a clear security warning about persistent external access, review cadence, least-privilege, or approval requirements. In an access-governance skill, normalizing permanent sharing can lead operators to create long-lived third-party access that is forgotten, hard to audit, and still active after business need ends.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The verification guide instructs users to run `CreateNamespaceAuth` and `DeleteNamespaceAuth` against real namespaces as part of routine validation, but it does not clearly warn that these commands change live access-control state. In an operational cloud environment, this can unintentionally grant another IAM principal access to a namespace or cause disruptive permission churn during testing, especially if users copy the commands into production with substituted values.

VirusTotal

59/59 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.