linode
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates infrastructure management through the
linode-cliutility andssh. It automates the creation of Linodes, firewall rules, and Kubernetes clusters using standard CLI operations. - [EXTERNAL_DOWNLOADS]: The skill references the installation of the official
linode-clitool from standard package repositories (PyPI, Homebrew). It also includes a documentation link to a Linode-managed GitHub IO page for Kubernetes autoscaling. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection due to the ingestion of user-defined parameters for infrastructure deployment.
- Ingestion points: Server labels, region codes, and instance types provided by the user in
SKILL.mdStep 0. - Boundary markers: No explicit delimiters or instructions to ignore embedded content in user data are present.
- Capability inventory: The skill executes shell commands via
linode-cli, performs remote management viassh, and writes configuration to a local.env.localfile. - Sanitization: User-provided inputs are not sanitized or validated before being interpolated into shell command arguments.
Audit Metadata