melech-pr-gardener

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

• Third-party content exposure detected (medium risk: 0.30). The agent skill monitors and processes GitHub pull requests, including review comments and issue interactions authored by outsiders, via gh pr list, gh pr view, and related CLI commands.

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

• Potentially malicious external URL detected (medium risk: 0.30). The routine configuration in references/ROUTINE.md instructs the agent to execute npx -y skills add AdirD/agent-shell-hamelech --skill melech-pr-gardener -g -y -a '*' to fetch and install the skill from a remote GitHub repository at runtime, representing a common/reputable unofficial operational dependency.