ccp

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). This skill explicitly downloads and installs items from public third-party repositories (e.g., "ccp find -r github " and "ccp install owner/repo" in SKILL.md), ingesting untrusted user-generated skills/agents/hooks into the hub that can materially alter agent behavior and enable indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill exposes runtime commands that fetch Git repositories (e.g., owner/repo or git@github.com:owner/repo.git via "ccp install" / "ccp source add") to install hub items (skills/agents/settings-templates) which are used as prompts/instructions or executable hooks, so external content fetched at runtime can directly control agent behavior.