skills/zjy365/seakills/sealos-deploy/Gen Agent Trust Hub

sealos-deploy

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates deployments by executing shell commands including docker, kubectl, and gh. These operations are integral to the deployment workflow. High-risk commands like kubectl delete are guarded by mandatory user confirmation rules in the primary instructions.
  • [EXTERNAL_DOWNLOADS]: During environment setup, the skill offers to install missing dependencies. For Docker installation on Linux, it utilizes the curl | sh pattern. This is an external code execution vector, but the skill is designed to ask the user for permission before running any installation scripts.
  • [PROMPT_INJECTION]: The skill ingests data from local project files such as README.md, package.json, and docker-compose.yml to determine deployment parameters. This presents a potential surface for indirect prompt injection if those files contain malicious instructions, though the risk is mitigated by the use of deterministic scripts for most parsing tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 11:26 AM