tempo

SUSPICIOUS. The skill’s purpose mostly matches its capabilities, and the installer appears to come from the official Tempo publisher. However, it relies on an unpinned pipe-to-shell installer, instructs the agent not to seek safer/alternate docs, and then entrusts an external CLI with wallet login, funding, credits spending, and card-related actions. This is not confirmed malware, but it is a medium-risk skill with meaningful supply-chain and financial-action trust concerns.

The code fragment itself does not perform file IO, network requests, or data processing directly, but it includes a suspicious AI invocation that attempts to coerce an external AI service into producing a tightly controlled output. This represents a potential supply-chain and data-leak risk if secrets are ever included in inputs or configurations are executed in user environments. The dominant concern is the AI prompt injection-like pattern and reliance on a third-party service in a package configuration, which could enable information leakage or unintended behavior in CI/CD pipelines.