processon-diagram-generator

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs a shell command at runtime to fetch and parse https://raw.githubusercontent.com/processonai/processon-skills/main/skills/processon-diagram-generator/version/github-version.json, and the fetched JSON (version/changelog) is used to control whether the agent must immediately interrupt and prompt the user—i.e., remote content fetched at runtime directly controls agent behavior.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill explicitly instructs the agent to run shell commands (node ./setup.mjs, a run_shell_command node -e fetch, and an npx ... -g update) and to write/clear user config at ~/.processon-diagram-generator/token.json, which causes the agent to modify the host machine's state (including potentially performing global installs) even though it does not request sudo or create new users.