document-release

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text can enter the LLM context via runtime-fetched PR/MR body content: the workflow reads the existing PR/MR body (gh pr view --json body / glab mr view ... description) and then appends/replaces a “## Documentation” section, so the agent ingests outsider-authored text from the PR/MR description.