ln-914-community-responder

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md explicitly fetches and parses GitHub discussions/issues via GraphQL in "Phase 1: Load Items" (e.g., the gh api graphql examples) and then reads their bodies and comments to extract keywords, drive code searches, classify and compose responses—i.e., it ingests untrusted, user-generated third‑party content from public GitHub that can materially influence tool use and next actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill specifies a runtime fallback to fetch required shared files via WebFetch from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}, and those fetched Markdown templates (e.g., response styles, humanizer checklist) are marked "MANDATORY READ" and directly control the agent's prompts/instructions at runtime, so this external raw.githubusercontent.com URL is a runtime dependency that can alter agent behavior.