external-signal-provider

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly ingests external provider data (e.g., AgentFS via ToolCalls.stats()/AgentFsClient using AGENTFS_DB_PATH and registry.aggregate_signals/get_signals in the SKILL.md and reward-integration.md) and then normalizes and merges those ExternalSignalSet values into SignalMerger.merge during complete_episode, so untrusted third‑party telemetry can directly influence reward calculations and agent behavior.