LOOM — Recursive Trust Graph on Canopy

LOOM is a SocialFi protocol where reputation is computed on-chain as a stake-weighted recursive trust graph. Every vouch is a staked directed edge and its pull on your score scales with your voucher's own earned trust. Sybils get nowhere because trust has to flow in from nodes that already have it. Trust isn't a score. It's a shape. ➡️ LINKS: ✅️ GitHub: https://github.com/0xkinno/loom ✅️ Live Demo: https://loom-snowy-pi.vercel.app/ --- Ten custom transaction types on Canopy: REGISTER_IDENTITY, WEAVE_THREAD, BREAK_THREAD, ENDORSE_SKILL, FORM_GUILD, JOIN_GUILD, POST_QUEST, SUBMIT_PROOF, ATTEST_CONTRIBUTION, CAST_WEIGHTED_VOTE Trust engine: stake-weighted PageRank (damping 0.85, 40 iterations) running inside the plugin's deliver handlers, recomputed on-chain after every topology change. Frontend screens: - The Weave — live force-directed trust graph, click any node for full identity breakdown - Vouch — stake CNPY to back someone on a specific skill - Guilds — trust-gated collectives with quest boards, proof submission, and peer(1] attestation - My Thread Map — shows exactly where each identity's trust comes from, backer by backer - Governance — trust-weighted voting, influence earned through the graph not token balance Stack: Canopy TypeScript plugin, Vite + React + d3-force, deployed on Vercel #canopy #crypto