alibabacloud-lakehouse-migration

SUSPICIOUS. The skill’s business purpose and capabilities are mostly coherent for schedule migration, and most file/API access is proportionate. The main risk is install/execution trust: it requires an externally unverifiable `aliyun-cli-lhm` plugin and a repo-local plugin installer, then uses that component in the credential-bearing Alibaba Cloud CLI path. That makes the overall footprint too trust-heavy for a routing/orchestration skill, even though there is no strong evidence of deliberate credential theft or overt malware.

SUSPICIOUS. The workflow purpose is coherent with the capabilities and the network destinations align with Alibaba Cloud APIs/OSS, so this is not confirmed malware. However, the skill depends on external CLIs/plugins whose specific provenance and release trail are not publicly verifiable here, and it forwards temporary credentials for OSS upload; that makes the skill high risk despite otherwise proportional data flows.

SUSPICIOUS. The skill’s core SQL-conversion purpose is coherent, but its actual footprint is broader than a local converter: it depends on an unverified external Alibaba CLI plugin, reads sensitive credential files, connects to target engines, and silently sends audit data off-box. The hidden audit flow and unverifiable plugin/install path make the skill medium-high risk even without evidence of confirmed malware.

SUSPICIOUS. The stated purpose and capabilities mostly align: an agent-driven schedule migration skill invoking LHM read/convert APIs, polling, and downloading result artifacts is coherent. Main concern is install/execution trust: the workflow depends on a specific aliyun CLI plugin (aliyun-cli-lhm) whose public official provenance could not be verified, even though the parent Alibaba Cloud CLI and plugin system are official. No clear credential exfiltration, attacker-controlled endpoint, or malicious pre-execution behavior is shown, but the unverifiable plugin dependency and transitive trust in a separate env skill keep risk elevated.

SUSPICIOUS. The skill's purpose and capabilities mostly align, and its network use fits cloud preflight checks, but it stores plaintext AK/SK locally and forwards credentials through an external CLI/plugin dependency whose specific `aliyun-cli-lhm` provenance was not verified. That combination makes the skill high-risk from a supply-chain and credential-handling perspective, though not confirmed malware.

SUSPICIOUS: the skill’s purpose and operations are coherent for template CRUD, and the data flow appears aligned to Alibaba/LHM APIs. However, it depends on a required `aliyun-cli-lhm` plugin whose public provenance could not be verified, so install trust is insufficiently established and overall risk is elevated despite otherwise proportionate scope.

The script is a deployment helper whose intended behavior is to configure a public Alibaba plugin index and install or update the lhm plugin. It contains no clear malicious payload or direct data theft behavior. However, always installing the latest remote plugin without version pinning, signature verification, or checksum validation creates a significant supply-chain risk: compromise of the public index, CDN, account, or plugin release could result in arbitrary code execution through the aliyun CLI. The fallback to an existing local plugin also weakens update assurance. Review and harden the trust model before use in sensitive environments.

SUSPICIOUS: the skill’s purpose and requested inputs are broadly consistent with LHM task automation, and no clear malicious or unrelated data exfiltration is shown. However, it depends on an external aliyun-cli-lhm plugin whose exact official provenance could not be verified, and that plugin may receive credentials/config from local files or environment variables, creating a high supply-chain and credential-forwarding risk.

SUSPICIOUS: The core behavior matches a cloud diagnostic skill, and Alibaba Cloud CLI usage is broadly coherent with the stated purpose. The main concerns are reliance on an unverified `lhm-common` executable and implicit credential-resolution features that can expand trust boundaries; no clear evidence of malicious exfiltration is present.

SUSPICIOUS. The core behavior matches the stated purpose, and Alibaba credential/API usage is broadly proportionate. However, the skill relies on internal, publicly unverifiable tooling and client components, so execution trust is not fully established; this raises supply-chain risk without clear evidence of malicious intent.