@6171/agent-change-ledger

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

• Third-party content exposure detected (medium risk: 0.30). The SKILL’s runtime workflow (“Workflow” steps 1–7) ingests only user-provided inputs such as “change request or issue text” and “files changed and their intended impact” and then emits a ledger, so outsider free text can be provided directly by the user to be processed by the agent.