lattice-spatial-collective

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

• Third-party content exposure detected (high risk: 0.70). The skill describes agents fetching and ingesting context shards from other peers via WebRTC/DHT lookups and IPFS-style CIDs and gossip (e.g., "Tier 2 shards requested on-demand via DHT lookup", "Agents share context updates with 3 random peers"), which clearly exposes the agent to untrusted, user-generated third-party content from the open peer network.