[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-98964":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":13,"contributorsCount":13,"subscribersCount":13,"size":13,"stars1d":13,"stars7d":13,"stars30d":13,"stars90d":13,"forks30d":13,"starsTrendScore":13,"compositeScore":15,"rankGlobal":10,"rankLanguage":10,"license":16,"archived":17,"fork":17,"defaultBranch":18,"hasWiki":19,"hasPages":19,"topics":20,"createdAt":10,"pushedAt":29,"updatedAt":30,"readmeContent":10,"aiSummary":10,"trendingCount":13,"starSnapshotCount":13,"syncStatus":31,"lastSyncTime":32,"discoverSource":33},98964,"pheromone-network","Wrste\u002Fpheromone-network","Wrste","A zero-dependency TypeScript memory kernel that gives any agent long-term memory. No training or backpropagation: each output reads a few tag-compatible neighbors, updated by a manual Hebbian rule. Links used often are reinforced; idle ones evaporate by half-life. Based on the paper Local Pheromone Network (arXiv:2606.30669).","",null,"TypeScript",83,0,81,36,"MIT License",false,"main",true,[21,22,23,24,25,26,27,28],"agent-memory","associative-memory","local-learning","memory","npm","offline-ai","semantic-recall","typescript","2026-10-04 05:19:52","2026-10-12 18:21:32",2,"2026-10-13 04:12:11","CREATED_QUERY"]