[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:omnigraph":3},"\u003Cp align=\"center\">\n  \u003Cpicture>\n    \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"assets\u002Fomnigraph-wordmark-dark.svg\">\u003C\u002Fsource>\n    \u003Cimg alt=\"OMNIGRAPH\" src=\"https:\u002F\u002Fraw.githubusercontent.com\u002FModernRelay\u002Fomnigraph\u002FHEAD\u002Fassets\u002Fomnigraph-wordmark.svg\" width=\"420\" \u002F>\n  \u003C\u002Fpicture>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Cstrong>Lakehouse graph database for context assembly & multi-agent coordination\u003C\u002Fstrong>\u003Cbr \u002F>\n  \u003Csub>Multimodal retrieval · Git-style branching · object-storage native\u003C\u002Fsub>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FModernRelay\u002Fomnigraph\u002Fblob\u002FHEAD\u002Fdocs\u002Fuser\u002Fquickstart.md\" rel=\"nofollow ugc noopener\">Quickstart\u003C\u002Fa>  · \n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FModernRelay\u002Fomnigraph\u002Fblob\u002FHEAD\u002Fdocs\u002Fuser\u002Fclusters\u002Findex.md\" rel=\"nofollow ugc noopener\">Docs\u003C\u002Fa>  · \n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FModernRelay\u002Fomnigraph-cookbooks\" rel=\"nofollow ugc noopener\">Cookbooks\u003C\u002Fa>  · \n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FModernRelay\u002Fomnigraph\u002Fblob\u002FHEAD\u002Fdocs\u002Fuser\u002Fcli\u002Freference.md\" rel=\"nofollow ugc noopener\">CLI\u003C\u002Fa>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FModernRelay\u002Fomnigraph\u002Fblob\u002FHEAD\u002FLICENSE\" rel=\"nofollow ugc noopener\">\u003Cimg alt=\"License: MIT\" src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-1b1b1f?style=flat-square&labelColor=1b1b1f\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fcrates.io\u002Fcrates\u002Fomnigraph-cli\" rel=\"nofollow ugc noopener\">\u003Cimg alt=\"crates.io\" src=\"https:\u002F\u002Fimg.shields.io\u002Fcrates\u002Fv\u002Fomnigraph-cli?style=flat-square&color=d71921&labelColor=1b1b1f\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FModernRelay\u002Fomnigraph\u002Fblob\u002FHEAD\u002Frust-toolchain.toml\" rel=\"nofollow ugc noopener\">\u003Cimg alt=\"Rust\" src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Frust-stable-1b1b1f?style=flat-square&labelColor=1b1b1f\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Chr \u002F>\u003Cp>Omnigraph is the operational state and coordination layer for fleets of agents.\u003Cbr \u002F>Run it as a server, declared as code; hundreds of agents operate and enrich the graph on parallel isolated branches, and every change is reviewed and merged safely.\u003C\u002Fp>\n\u003Cp>Join the \u003Ca href=\"https:\u002F\u002Fjoin.slack.com\u002Ft\u002Fomnigraphworkspace\u002Fshared_invite\u002Fzt-3wfpglyxj-lHvJGhuySPfqLtN35uJZNw\" rel=\"nofollow ugc noopener\">Omnigraph Slack community\u003C\u002Fa>\nto ask questions, share feedback, and follow development.\u003C\u002Fp>\n\u003Ch2>Key capabilities\u003C\u002Fh2>\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>Capability\u003C\u002Fth>\n\u003Cth>What it gives you\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>\u003Cstrong>Declared as code\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>A \u003Ccode>cluster.yaml\u003C\u002Fcode> declares graphs, schemas, stored queries, embedding providers, and policies; \u003Ccode>cluster apply\u003C\u002Fcode> converges it and \u003Ccode>omnigraph-server\u003C\u002Fcode> brings every graph online at \u003Ccode>\u002Fgraphs\u002F{id}\u002F…\u003C\u002Fcode>.\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Built for fleets of agents\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Hundreds of agents enrich the graph on \u003Cstrong>parallel isolated branches\u003C\u002Fstrong>; changes are reviewed and merged safely, Git-style, across the whole graph.\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Multimodal retrieval\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Graph traversal + vector ANN + full-text + Reciprocal Rank Fusion in \u003Cstrong>one\u003C\u002Fstrong> query runtime, for context assembly.\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Security as code\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Cedar policy enforced \u003Cstrong>server-side on every mutation\u003C\u002Fstrong>, per-graph and server-wide; bearer auth; actor\u002Faudit tracking.\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Runs on your infrastructure\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Any S3-compatible object store: \u003Cstrong>on-prem via RustFS \u002F MinIO\u003C\u002Fstrong>, or AWS S3 \u002F R2 \u002F GCS. VPC, on-prem, hybrid; your data never leaves your store.\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Open, versioned storage\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flance-format\u002Flance\" rel=\"nofollow ugc noopener\">\u003Ccode>Lance\u003C\u002Fcode>\u003C\u002Fa> columnar format: branchable, time-travelable, with native blob-as-data (docs, images, video).\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n\u003Ch2>What you can build\u003C\u002Fh2>\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>Use case\u003C\u002Fth>\n\u003Cth>What it's for\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>\u003Cstrong>Company brain\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Org knowledge unified into one graph every agent can query\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Agentic memory\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Durable, versioned memory: a branch per agent or per task, merged on review\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Context graph\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Decision traces and codified tribal knowledge for retrieval\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Dev graph\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Issues & dependency model that coding agents read and write\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>R&D \u002F ML data layer\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>Experiments and trials written into branches, versioned for training & eval\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n\u003Ch2>Install\u003C\u002Fh2>\n\u003Cpre>\u003Ccode class=\"language-bash\">curl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002FModernRelay\u002Fomnigraph\u002Fmain\u002Fscripts\u002Finstall.sh | bash\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>This installs \u003Ccode>omnigraph\u003C\u002Fcode> (CLI) and \u003Ccode>omnigraph-server\u003C\u002Fcode> into \u003Ccode>~\u002F.local\u002Fbin\u003C\u002Fcode> from\npublished release binaries. Or with Homebrew:\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-bash\">brew tap ModernRelay\u002Ftap\nbrew install ModernRelay\u002Ftap\u002Fomnigraph\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Ch2>Set it up with an AI agent\u003C\u002Fh2>\n\u003Cp>Omnigraph is built to be run by coding agents. Two ways in:\u003C\u002Fp>\n\u003Cp>\u003Cstrong>Teach your agent the playbook.\u003C\u002Fstrong> This repo ships the\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FModernRelay\u002Fomnigraph\u002Fblob\u002FHEAD\u002Fskills\u002Fomnigraph\" rel=\"nofollow ugc noopener\">\u003Cstrong>\u003Ccode>omnigraph\u003C\u002Fcode> agent skill\u003C\u002Fstrong>\u003C\u002Fa>: the operational playbook\ncovering cluster mode, the two config surfaces, schema evolution, query linting,\ndata writes, branches, Cedar policy, and the common gotchas.\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-bash\">npx skills add ModernR\n\u003C\u002Fcode>\u003C\u002Fpre>\n",1784240407753]