chronicle-read

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

• Third-party content exposure detected (high risk: 0.70). The required workflow reads narrative chapter files from /chronicle/ (e.g., read /chronicle/NNNNNN.md), and those chapters are auto-maintained from /chat_history/ chatter, which can include free-form messages authored by outsiders; thus outsider-authored text can be ingested into the LLM via the chronicle context auto-prepends.