setup-matt-pocock-skills

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly configures agent workflows to fetch and act on GitHub issues (see issue-tracker-github.md and SKILL.md instructions such as "Run gh issue view <number> --comments"), which are untrusted, user-generated third-party content the agent is expected to read and whose contents can materially influence triage/labeling and other tool-driven actions, enabling indirect prompt injection.