copilot-coding-agent

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

• Third-party content exposure detected (high risk: 0.90). This skill's workflow (SKILL.md "Issue created/labeled" → "Copilot Coding Agent starts processing the issue" and the usage examples showing gh issue create --body "...", plus the "External system integration" note) auto-assigns Copilot to read and act on user-generated GitHub issue bodies (including issues created via Jira/Zapier), so untrusted third-party content can directly influence the agent's actions.