gitea-pagerank-workflow

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

• Third-party content exposure detected (high risk: 0.80). The skill's Agent Workflow and references/agent-coordination.md/CLAUDE.md explicitly instruct the agent to call gitea-robot, tea, and curl against the Gitea API (e.g., fetching issue bodies, comments, and notifications from https://git.terraphim.cloud), which are user-generated/untrusted content and are read and acted on (claiming work, reserving paths, deciding next actions), so third-party content can materially influence agent behavior.