{"_id":"@emergent-wisdom/meaning-model-mcp","_rev":"14-48c388e2f62d37309bb6aa8394a71949","name":"@emergent-wisdom/meaning-model-mcp","dist-tags":{"latest":"0.6.8"},"versions":{"0.1.0":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.1.0","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.1.0","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"26ca821601be09561ea6080c50e2eab38845916a","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.1.0.tgz","fileCount":54,"integrity":"sha512-gwreml8NCXbHd3/+X7ovthH/Bre2RyQWv5RAB+E3ry7je9GQAm4fp1tbw1LjpNAZRBF+6r7WDjIC9JHHKfwsmA==","signatures":[{"sig":"MEUCIQCiRwgflVCEaUau7i4bzJiPC/+DyD++/vDqvVcZHM5OaQIgfZvW5iD5aDf7PC0zgLam7vPOLa0XEuOnq0YUgu9Wuqg=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":8204657},"type":"module","_from":"file:emergent-wisdom-meaning-model-mcp-0.1.0.tgz","engines":{"node":">=22.18.0"},"scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/mcp-releases/meaning-model-mcp-0.1.0/emergent-wisdom-meaning-model-mcp-0.1.0.tgz","_integrity":"sha512-gwreml8NCXbHd3/+X7ovthH/Bre2RyQWv5RAB+E3ry7je9GQAm4fp1tbw1LjpNAZRBF+6r7WDjIC9JHHKfwsmA==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server with a Rust engine and versioned modeling resources","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.1.0_1788567442201_0.6275640748283047","host":"s3://npm-registry-packages-npm-production"}},"0.1.1":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.1.1","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.1.1","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"e2cafd01f5c69d8a7f2fbf213a6d6270912b8134","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.1.1.tgz","fileCount":55,"integrity":"sha512-X0k9HdiwKxMRGakqi3Jc6Rd+nqgmWJXBU8DQvxGTPFXg2SNFiXZnpy5/cBfU71XMTd5k/3JmHMVmlAKwDp9U2A==","signatures":[{"sig":"MEYCIQCdiHkX6HWjiH6PBu9ar9liUfyShbbcBTSiry175V+nBQIhAK98lFmhzowyiAp1rY+iKpQirl8R2/O46ht194/CeC1J","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":8209713},"type":"module","_from":"file:emergent-wisdom-meaning-model-mcp-0.1.1.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/mcp-releases/meaning-model-mcp-0.1.1/emergent-wisdom-meaning-model-mcp-0.1.1.tgz","_integrity":"sha512-X0k9HdiwKxMRGakqi3Jc6Rd+nqgmWJXBU8DQvxGTPFXg2SNFiXZnpy5/cBfU71XMTd5k/3JmHMVmlAKwDp9U2A==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server with a Rust engine and versioned modeling resources","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.1.1_1788645158523_0.6151240700515057","host":"s3://npm-registry-packages-npm-production"}},"0.2.1":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.2.1","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.2.1","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"61c85ba4f5e7b899093015cd1d9da5016bf571fb","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.2.1.tgz","fileCount":60,"integrity":"sha512-ceswYvQCufOc4coGKBkibYxF7+zjcibm4BEq7cQkWSD/CtWZYiKyvw4dXZDnIGpjeZ+7BN6yzR/nqiPZoFN5Tw==","signatures":[{"sig":"MEUCIQCCgARcXnwfkkDDMYZhRTBmTDQgUyLMfLf/JrefIm34SQIgZcMo/bopr0AK03w4R3rNgPr6Miyf9d1FI1LuqevCT0c=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":8340437},"type":"module","_from":"file:build/npm-package-Pusvy2/emergent-wisdom-meaning-model-mcp-0.2.1.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-publish-2026-09-09/build/npm-package-Pusvy2/emergent-wisdom-meaning-model-mcp-0.2.1.tgz","_integrity":"sha512-ceswYvQCufOc4coGKBkibYxF7+zjcibm4BEq7cQkWSD/CtWZYiKyvw4dXZDnIGpjeZ+7BN6yzR/nqiPZoFN5Tw==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server with a Rust engine and versioned modeling resources","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.2.1_1789200321064_0.09360225659572086","host":"s3://npm-registry-packages-npm-production"}},"0.3.0":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.3.0","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.3.0","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"394697d84403104386d1cfc4c8aacfcdab003a62","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.3.0.tgz","fileCount":110,"integrity":"sha512-dz6R34OXMCgEc47Cyxr8MGdpgQM1tM6p5abeMhNVpJMadd+gOK8iKGtzXXf0/rXBcMCgu9jInGVC8kccrLhkjg==","signatures":[{"sig":"MEUCIFL1WYpy1xWDBgh0adGUVtNNNP7pVPfBnrbVq4NMvBqGAiEAx03/rpdorEmrVZNRvnZFML31TyM35bv9ehwxe7gMZew=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIAd/3/WtDRJZ+ienz1U4Z0cxGVTAB3LsrlUiQhlx+REoAiEArKk6w5upDFgTzhAB0BpJEWvB4wRUCo2wO3vTic7AYR0=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":9683867},"type":"module","_from":"file:/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-publish-2026-09-09/build/npm-package-79dyi7/emergent-wisdom-meaning-model-mcp-0.3.0.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-publish-2026-09-09/build/npm-package-79dyi7/emergent-wisdom-meaning-model-mcp-0.3.0.tgz","_integrity":"sha512-dz6R34OXMCgEc47Cyxr8MGdpgQM1tM6p5abeMhNVpJMadd+gOK8iKGtzXXf0/rXBcMCgu9jInGVC8kccrLhkjg==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.3.0_1790358550539_0.2691179917211899","host":"s3://npm-registry-packages-npm-production"}},"0.5.0":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.5.0","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.5.0","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"d082d6832fd7ae4cd45817f94803d91c2e9bbb7b","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.5.0.tgz","fileCount":170,"integrity":"sha512-jfEO90D/T2WmTaDRDXxCuSRrB5iiaQ6VoFkQCmua51J0CMGtIVgY/t72bCSr+/m/MMpM0UVySs5BAzkkK4V0ow==","signatures":[{"sig":"MEYCIQCfMZrsgW4pdM+HrNsq4lkwWerrVmlMdxTDWf2FuVQy0AIhAPeCKOmKFGI2eB81y+49Vcf3zdEEe6tnwROFpbW7KweT","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIQCwSJMOKUla2zvQEyRQ//FwtQf0jNaarMIeDBeNnF8BIgIgXLEDGnVxNEhAZsMZS06+00Zl3doD4jJyB4lf7bcr8PU=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":12883595},"type":"module","_from":"file:build/npm-package-lln7Yj/emergent-wisdom-meaning-model-mcp-0.5.0.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-release-0.5.0/build/npm-package-lln7Yj/emergent-wisdom-meaning-model-mcp-0.5.0.tgz","_integrity":"sha512-jfEO90D/T2WmTaDRDXxCuSRrB5iiaQ6VoFkQCmua51J0CMGtIVgY/t72bCSr+/m/MMpM0UVySs5BAzkkK4V0ow==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.5.0_1790537553506_0.9812572185313615","host":"s3://npm-registry-packages-npm-production"}},"0.5.2":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.5.2","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.5.2","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"d0229d86d00ba57ebf47fa09e93cb83d2748d67a","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.5.2.tgz","fileCount":175,"integrity":"sha512-fbFU3VomXeZ5K1rvvBl279wh6h4vkyaUNm8ouB3V+vEJV4XGviFKoLzprkTxdxpRdFuRpLhwjHyf7D6K5wcvkA==","signatures":[{"sig":"MEQCIFmmTT8tW+BNYocF4WiAyqsuZC0VSJHme8bvmvruCOGCAiBq39PEt/tE0eGs5tzRs5o/Vf23GFdJCTsyYp2ijlNYQQ==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIQDP80BTqEcEMYBxP+K8l+xMrkSMZI+B02LaMc6v8Hn6aAIgFXs9QkZYWRZAGId0l3fhSGIALSEZ2PWpIRH3kxv/DGg=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":12946834},"type":"module","_from":"file:build/npm-package-qSnLs3/emergent-wisdom-meaning-model-mcp-0.5.2.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-release-0.5.1/build/npm-package-qSnLs3/emergent-wisdom-meaning-model-mcp-0.5.2.tgz","_integrity":"sha512-fbFU3VomXeZ5K1rvvBl279wh6h4vkyaUNm8ouB3V+vEJV4XGviFKoLzprkTxdxpRdFuRpLhwjHyf7D6K5wcvkA==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.5.2_1790595550970_0.758455996601725","host":"s3://npm-registry-packages-npm-production"}},"0.6.0":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.6.0","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.6.0","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"6986efdcad6f6c2e317a3c22987f23cc1aa81cac","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.0.tgz","fileCount":183,"integrity":"sha512-uw6PQc4hHY/gDPUUAuilECnw8dGK+VcWVWqrbUw/GKF3ejgQMAPrWFvLzYNVZ7eYyaeSfA4q9YQvQFrqH66tzA==","signatures":[{"sig":"MEUCIHeuafpPs6tiM6TpBrA3U2hIYQt0oFUVWn6qVSvyUuBeAiEA53QSFEWuPsdlxKEV4qbXL2Bpn2KX+Io3QTaGVlLEcYI=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIBGHEekMT+rcKerQrbTWbEK6AvtSrbreA2GFHsF8b5UOAiEAm7KRgdDzJlhsAr2SRElk2vtysKvyxi2CbbHHWWO5Hl8=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":13032605},"type":"module","_from":"file:build/npm-package-qUbe1D/emergent-wisdom-meaning-model-mcp-0.6.0.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-release-0.5.1/build/npm-package-qUbe1D/emergent-wisdom-meaning-model-mcp-0.6.0.tgz","_integrity":"sha512-uw6PQc4hHY/gDPUUAuilECnw8dGK+VcWVWqrbUw/GKF3ejgQMAPrWFvLzYNVZ7eYyaeSfA4q9YQvQFrqH66tzA==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.6.0_1790711417391_0.6493116769216154","host":"s3://npm-registry-packages-npm-production"}},"0.6.1":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.6.1","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.6.1","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"ac512ebb63ad0ff52d694aa0a500cfcf9b3173cc","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.1.tgz","fileCount":185,"integrity":"sha512-fnM21i+ppuB0G/AKcCea1f/gvGyrVLQaEKLBt8Jiu8Gl5ijpPR1Y747SDjwO8LgtvIcYWrSo5aTzlAjhGrASJg==","signatures":[{"sig":"MEUCIQCxy5FWD02CFSH6PIOtIbfZRQdPx4oUADg5pmBQmxIJyAIgGRA71Wy2nfoDyT2GQAAsfOwLvObGhzrE//7rWUPRc9k=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIFgcGv/nAa6Bcd4owANsmc2AS2oyCx9W4qb77tZ9GCbEAiEA+G3F7Ax/9z+9AcEHeA/ZBEkzNB+Cvyj6632FBv2IfUE=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":13092176},"type":"module","_from":"file:/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-release-0.6.1/build/npm-package-BpxJQ7/emergent-wisdom-meaning-model-mcp-0.6.1.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-release-0.6.1/build/npm-package-BpxJQ7/emergent-wisdom-meaning-model-mcp-0.6.1.tgz","_integrity":"sha512-fnM21i+ppuB0G/AKcCea1f/gvGyrVLQaEKLBt8Jiu8Gl5ijpPR1Y747SDjwO8LgtvIcYWrSo5aTzlAjhGrASJg==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.6.1_1790812297349_0.8117980435938132","host":"s3://npm-registry-packages-npm-production"}},"0.6.2":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.6.2","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.6.2","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"13fd2c17dba6865c9a1b5c4da8fd5bda8fea551d","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.2.tgz","fileCount":185,"integrity":"sha512-SKBVaS1mDoYWsVKKFqaYQ5cVA864GvJDl1KRutJmnGchZRoEHMyoegu/avzrgbLDyFX0ceWFfQe1O3xP8GJIUQ==","signatures":[{"sig":"MEYCIQCS9mOB8fVHYfeE4fIOnCB3ByZkkCq5K4j/wCZvICYEwwIhALDAQIkohbq+TBFbFe/+4oavUgX4MaBWjcObKTtO4QCc","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEYCIQDgIqhcPidC7Q9gGgKRACHfFAIxAvxbMuJaSpNHVg8deAIhAOqM/Xa522GNVLhcl4vuWKwKeHv5cGsXfDPMw5UH54ff","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":13107353},"type":"module","_from":"file:build/npm-package-OlEqY9/emergent-wisdom-meaning-model-mcp-0.6.2.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-release-0.6.1/build/npm-package-OlEqY9/emergent-wisdom-meaning-model-mcp-0.6.2.tgz","_integrity":"sha512-SKBVaS1mDoYWsVKKFqaYQ5cVA864GvJDl1KRutJmnGchZRoEHMyoegu/avzrgbLDyFX0ceWFfQe1O3xP8GJIUQ==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.6.2_1790819645084_0.25307115552440607","host":"s3://npm-registry-packages-npm-production"}},"0.6.3":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.6.3","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.6.3","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"0de70a7fd1d4d0b2f277f4eb73e4d626ae4ff2db","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.3.tgz","fileCount":185,"integrity":"sha512-GhR3Ll1+63oeqKeFsvH6/cq4sfVuKy3g38WaMxolBE8yKiunv9MRjeNNeWETY6zzYTFb/NfRXUSy0dJNg8HotQ==","signatures":[{"sig":"MEUCIAstqjAf2+gZmtMV27VsR8spAfqLju9gAePM9iqzaP0uAiEAl/1GL9me5Y1t7wk/1tzK4c1VrUQ43thl6wmo6cCleME=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIFdHORAPioku2Hrb3rnmJBPflkdKvC70Qa6vQFGsbM9jAiEA3cQR4Ce9GJ0Gupka9+ijO72TVpaSPpMgnjg9AvoRAO4=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":13114064},"type":"module","_from":"file:build/npm-package-PIk1tw/emergent-wisdom-meaning-model-mcp-0.6.3.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-release-0.6.1/build/npm-package-PIk1tw/emergent-wisdom-meaning-model-mcp-0.6.3.tgz","_integrity":"sha512-GhR3Ll1+63oeqKeFsvH6/cq4sfVuKy3g38WaMxolBE8yKiunv9MRjeNNeWETY6zzYTFb/NfRXUSy0dJNg8HotQ==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.6.3_1790831504164_0.13842702302057974","host":"s3://npm-registry-packages-npm-production"}},"0.6.5":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.6.5","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.6.5","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"0da0b7e52028eb9c2d9f4b30712ce7fc8bdd2036","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.5.tgz","fileCount":185,"integrity":"sha512-ZCIue1/2LI/Nk5+dn5DeRnW9UoCi/uJwFlA6XQgKUHd/d2ZGxR85nihsjrbtjG7vYoaXdfIyQx/iEXOve/iNFQ==","signatures":[{"sig":"MEQCIEexJbEEGzKOWxMXV0bxFzuB0V3AnxP4dFQ9aKEkVozxAiBQaWAUC6JOkOUI0ADWMjyrjnTZyor34OPoS77mqUW0wQ==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIA20GHv2t0zS+isY56PLa+T4EU8daLAluC06PNle6+8KAiEAqMt4k3kFrKC3kuZB8RDwyuRc5y6v1UPaP9xIUlc1hGA=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":13129243},"type":"module","_from":"file:build/npm-package-ycNKcf/emergent-wisdom-meaning-model-mcp-0.6.5.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-viewer-order-2026-10-02/build/npm-package-ycNKcf/emergent-wisdom-meaning-model-mcp-0.6.5.tgz","_integrity":"sha512-ZCIue1/2LI/Nk5+dn5DeRnW9UoCi/uJwFlA6XQgKUHd/d2ZGxR85nihsjrbtjG7vYoaXdfIyQx/iEXOve/iNFQ==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.6.5_1790908838932_0.8661002915398208","host":"s3://npm-registry-packages-npm-production"}},"0.6.6":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.6.6","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.6.6","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"a4955365a15222214275902831923015ac20c0f7","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.6.tgz","fileCount":189,"integrity":"sha512-OPPSOxJOCStG9SIon+d70tPOIY2w1B6mCpvb0qBcfQI+gIrlKnSHxWu8wVEMPW0mmlZWYlacUdRN9NJOGHABmg==","signatures":[{"sig":"MEUCIDq37lKXTWRyeRLEIMEzkedXimNbYeroix7/QCvANRZGAiEA5tsqQpru0F33cCHCccqU2NEADcX/dUUhN3bcLT1GD/8=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIQC8FwekdVwYbdyTx4vc+rf6vGR8iAuOQ/QdEXbk9KR7zwIgLTIlSrCIJqoWKYgRKyXutKfR9MlSePCbhR3aCcfT53U=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":13255942},"type":"module","_from":"file:build/npm-package-MKhN3H/emergent-wisdom-meaning-model-mcp-0.6.6.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-viewer-order-2026-10-02/build/npm-package-MKhN3H/emergent-wisdom-meaning-model-mcp-0.6.6.tgz","_integrity":"sha512-OPPSOxJOCStG9SIon+d70tPOIY2w1B6mCpvb0qBcfQI+gIrlKnSHxWu8wVEMPW0mmlZWYlacUdRN9NJOGHABmg==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.6.6_1791002307208_0.929490949538085","host":"s3://npm-registry-packages-npm-production"}},"0.6.7":{"name":"@emergent-wisdom/meaning-model-mcp","version":"0.6.7","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","_id":"@emergent-wisdom/meaning-model-mcp@0.6.7","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"dist":{"shasum":"fd2be1d3b9349f7d90d10745da9ec942f45df935","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.7.tgz","fileCount":189,"integrity":"sha512-kPx64ASd5wS/Y7iQV7EqSCHxsudJrGS70C6nAUGeJOtNCzCb1Py2SQhZUMnBbgN40G4Gdta5H7nqV3td2U1SWw==","signatures":[{"sig":"MEYCIQC8KE/mqpUP5mrzrJHMP7i7J8DZgFRsYDG2BICsVf5W/AIhAITfMbrha2VuA0OhlO26JAWRHmc8GBF/nw8ki3IRTIEn","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEUCIG+dBozGJBbOax1y9eb9Tr1sq8mApcqfvzcoxUi1OXl1AiEA5B2dCGUSqvl5VXGYfHOo6rnuSNTkgw2QYA3am/BFh+Y=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":13263753},"type":"module","_from":"https://github.com/emergent-wisdom/meaning-model/releases/download/v0.6.7/emergent-wisdom-meaning-model-mcp-0.6.7.tgz","engines":{"node":">=22.18.0"},"mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"_resolved":"https://github.com/emergent-wisdom/meaning-model/releases/download/v0.6.7/emergent-wisdom-meaning-model-mcp-0.6.7.tgz","_integrity":"sha512-kPx64ASd5wS/Y7iQV7EqSCHxsudJrGS70C6nAUGeJOtNCzCb1Py2SQhZUMnBbgN40G4Gdta5H7nqV3td2U1SWw==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/meaning-model-mcp_0.6.7_1791061618099_0.626606483007865","host":"s3://npm-registry-packages-npm-production"}},"0.6.8":{"_id":"@emergent-wisdom/meaning-model-mcp@0.6.8","bin":{"meaning-model-mcp":"mcp-server/bin/meaning-model-mcp.mjs"},"bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"dist":{"shasum":"4e66250de662b3bb6c141f49d14178214be80f98","tarball":"https://registry.npmjs.org/@emergent-wisdom/meaning-model-mcp/-/meaning-model-mcp-0.6.8.tgz","fileCount":190,"integrity":"sha512-2srOKZ/hoEqc7/kZSXhwbv0P3Ma2eVvTmuatM7bxgwlTutS4ntpYUnWmTFEWFkCei1crlqicqsnRLzAlEdA3rQ==","signatures":[{"sig":"MEYCIQCGI5ruJw/FsgWP50I6pRVsD6WxweoBWg2SoL2yp/yycQIhAJydgs5kdMJefgNabAha5T6sWEub6r5DLJygkuHK68ow","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCIGGjZRRxdVYHfpS3mQ484PqCCZb0n+ElpWoUJM3X+T4sAiEAkxn4iC9L7o8si01PjPbbcqf7AheQQhBkGweK1jUH7No="}],"unpackedSize":13291056},"name":"@emergent-wisdom/meaning-model-mcp","type":"module","_from":"file:build/npm-package-1obtEv/emergent-wisdom-meaning-model-mcp-0.6.8.tgz","author":{"name":"Henrik Westerberg"},"engines":{"node":">=22.18.0"},"license":"(MIT AND CC-BY-4.0)","mcpName":"io.github.emergent-wisdom/meaning-model","scripts":{"start":"node mcp-server/bin/meaning-model-mcp.mjs","build:engine":"node mcp-server/bin/meaning-model-mcp.mjs --build-engine","install:engine":"node mcp-server/bin/meaning-model-mcp.mjs --install-engine","verify:resources":"node scripts/verify-resources.mjs"},"version":"0.6.8","_npmUser":{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"},"homepage":"https://github.com/emergent-wisdom/meaning-model#readme","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"_resolved":"/Users/parcel/Documents/Code/Emergent Wisdom/meaning-model-alien-fixes-2026-10-04/build/npm-package-1obtEv/emergent-wisdom-meaning-model-mcp-0.6.8.tgz","_integrity":"sha512-2srOKZ/hoEqc7/kZSXhwbv0P3Ma2eVvTmuatM7bxgwlTutS4ntpYUnWmTFEWFkCei1crlqicqsnRLzAlEdA3rQ==","repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.12.1","description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","directories":{},"maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"_nodeVersion":"26.0.0","dependencies":{"zod":"4.4.3","@modelcontextprotocol/server":"2.0.0"},"publishConfig":{"access":"public"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/meaning-model-mcp_0.6.8_1791146245119_0.10224155430588944"}}},"time":{"created":"2026-09-05T00:17:22.010Z","modified":"2026-10-04T20:37:25.422Z","0.1.0":"2026-09-05T00:17:22.405Z","0.1.1":"2026-09-05T21:52:38.722Z","0.2.1":"2026-09-12T08:05:21.226Z","0.3.0":"2026-09-25T17:49:10.654Z","0.5.0":"2026-09-27T19:32:33.645Z","0.5.2":"2026-09-28T11:39:11.107Z","0.6.0":"2026-09-29T19:50:17.561Z","0.6.1":"2026-09-30T23:51:37.461Z","0.6.2":"2026-10-01T01:54:05.234Z","0.6.3":"2026-10-01T05:11:44.283Z","0.6.5":"2026-10-02T02:40:39.070Z","0.6.6":"2026-10-03T04:38:27.361Z","0.6.7":"2026-10-03T21:06:58.208Z","0.6.8":"2026-10-04T20:37:25.255Z"},"bugs":{"url":"https://github.com/emergent-wisdom/meaning-model/issues"},"author":{"name":"Henrik Westerberg"},"license":"(MIT AND CC-BY-4.0)","homepage":"https://github.com/emergent-wisdom/meaning-model#readme","keywords":["mcp","mcp-server","model-context-protocol","meaning-model","modeling","simulation","rust","storytelling","ideation"],"repository":{"url":"git+https://github.com/emergent-wisdom/meaning-model.git","type":"git","directory":"mcp-server"},"description":"Meaning Model MCP server: revisable general-purpose models with opt-in storytelling and ideation add-ons, on a Rust engine","maintainers":[{"name":"henrikwesterberg","email":"henrik.westerberg@emergentwisdom.org"}],"readme":"# Meaning Model MCP server\n\n`@emergent-wisdom/meaning-model-mcp` exposes the Meaning Model's Rust engine\nthrough MCP over stdio. The package includes JavaScript, Rust source and its\nlockfile, modeling profiles, versioned paper resources, and checked presets.\nIt contains no prebuilt engine or build cache.\n\nThe calling LLM builds an explicit world model from its learned knowledge and\navailable evidence. It develops processes, relationships and concepts, then\nuses those structures for further inference and abstraction. **Progressive\nresolution** opens detail under existing commitments; the **shared construction\nmedium** connects the world, its interpretations and documents for coordinated\nrevision. Inferred accounts remain distinct from observations.\n\nComparative judgments can be made explicit through a Cut: one declared unit is\ndivided under a question among exclusive answers and an explicit remainder.\nMeasurements retain their external units.\n\nStart with [installation](#install), connect the server to your assistant, and\nask it to begin. The assistant calls `life_modeling_context` for purpose-specific\noperational guidance. Reading the papers is optional.\n\nSee the [changelog](https://github.com/emergent-wisdom/meaning-model/blob/main/CHANGELOG.md) for release changes and upgrade notes.\n\n## Open your model\n\nVersion 0.6.8 includes the browser viewer in the MCP package.\n\nAfter installing this MCP and its matching engine, connect it to your assistant as\nusual. Ask **“Open the model we are working on”** or **“Show me this story.”** The\nassistant calls `life_model_viewer_open` with the model or graph revision and returns\na local link. Open that link in your browser. No separate viewer installation,\nGitHub checkout, run transcript or website account is needed.\n\nWhile writing, ask **“Keep the viewer following as we work.”** The assistant\nuses `mode: \"live\"` with the story's `graphHash`. Saved revisions update the\nbound model and prose at the same URL through guarded page refreshes, retaining\nreading position, view, record, time and camera. Updates wait while you interact\nor play; branches or insufficient access pause following. This follows saved\nrevisions, not unfinished tokens. Exact snapshots remain the default, and\nmodel-only links stay fixed. Opening a local live viewer does not update the\npublic website or publish private drafts.\n\nThe same **Processes / Tree / Terrain** viewer opens models with recorded time\npositions. Events can be displayed without numeric curves; curves appear only\nwhere the model supplies them. Models without time positions remain inspectable\nin Graph mode without an invented clock.\n\nAsk **“Open these models together”** to get a **Model** selector in the viewer.\nThe assistant supplies the other selections through `additionalModels`, each\nwith its own scopes. First visits keep the view and fit the selected model’s\ntime range. Returning to a snapshot in the same tab restores its own view, time\nand selection. Switching models loads that snapshot. Only models explicitly\nopened together appear in this selector.\n\nThe alternative **Graph mode** displays each saved revision's native records.\nFilter record types, find a record by name or ID, and click it to inspect\nits declared links and full record. **Neighbors only** focuses its connections;\n**Recenter** restores an overview. Positions are a layout, not time or measured\nvalues. Exact snapshot links stay fixed; live graph links follow saved revisions.\n\nWhen prose is present, **Read full document** opens the complete manuscript.\n**Read this passage** opens it at a selected passage; **Close the story** returns to\nexploration. Selecting a record keeps the reader closed. **Structure** offers an\nexpandable record view. **Show it as** switches among Processes, Tree, Terrain,\nGraph, Structure and Space on the same page, retaining the record selection and time.\n**Coarse view** returns to a broad overview; more detail opens subprocesses.\n**Recenter** (or Home outside text fields) fits the current representation.\n**Documents & notes** sits with the top controls. **Hide controls** clears that\ntoolbar while **Show controls** remains available.\n**Space** draws declared coordinates in their reference frame, lists qualitative\nsettings, and distinguishes dated home or work bases from continuous presence\nand observed travel. Graph and Space use the same movement keys as the time\nviews: W/A/S/D along the ground, Q/E vertically, arrows to look, and Shift for\nspeed. Text fields keep their normal typing behavior.\nThe original viewer's reading-position track highlights document links without changing world time;\nits reading buttons also open the complete manuscript. Optional Document spans\nfollow authored passage boundaries after edits and display unresolved boundaries.\n\nThe browser and MCP must run on the same computer. The link lasts while that MCP\nprocess runs; its sixteen most recently opened views remain available. This is a\ncomplete author view, requiring the supplied `accessScopes` to cover every scoped\nrecord in the exported model and graph. Partial access is refused rather than\npresented as a complete model. Opening a viewer does not publish the model.\n\n## Workflows over one model\n\nOne MCP server supports these workflows over the same engine and graph:\n\n- **Modeling.** General-purpose, revisable world models of whatever you choose\n  to model, such as a market, an institution or a technology, with an optional\n  Jev estimator for cheap first estimates. See\n  [general-purpose modeling](#general-purpose-modeling-and-optional-jev-estimation).\n- **Narration.** The storytelling add-on writes fiction from a model, with an\n  author model, whole-life character trends, scene review, alignment audits and\n  deepening passes. See the [storytelling add-on guide](https://github.com/emergent-wisdom/meaning-model/blob/main/profiles/STORYTELLING_ADDON.md).\n- **Human authorship with feedback.** The human writes and decides; the LLM\n  supplies feedback. Enable `MEANING_MODEL_ADDONS=storytelling` to use\n  `life_story_feedback`, which prepares a read-only task from supplied text or\n  an exact project revision. No existing world or fictional\n  author persona is required for feedback on an excerpt. See the\n  [feedback guide](https://github.com/emergent-wisdom/meaning-model/blob/main/docs/HUMAN_AUTHOR_FEEDBACK.md).\n- **Agent and user memory.** Model ongoing work and reported personal history,\n  with preferences, goals, decisions, beliefs and learning linked across time.\n  Capture useful records within the chosen scope and recover them across sessions\n  when durable storage is configured. See the [memory guide](https://github.com/emergent-wisdom/meaning-model/blob/main/docs/MEMORY.md).\n- **Ideation.** The alien add-on searches for solution mechanisms through\n  invented worlds, following *Ontology of the Alien*, and curates them into a\n  revisable map of idea families. See the [alien add-on guide](https://github.com/emergent-wisdom/meaning-model/blob/main/profiles/ALIEN_ADDON.md).\n\nModeling and memory are always available; the two add-ons are opt-in with\n`MEANING_MODEL_ADDONS`. Each workflow keeps what it does as revisable,\ninspectable structure. No workflow turns an estimate into an observation, a review\ninto an accepted fact, or an idea into evidence that it works. All share\nmacro-to-micro exploration, recursive discovery and attributed Understanding\nNodes. Their purpose and the human's delegation determine what the agent may\ninfer, record or change.\n\nIn memory work, the assistant begins with the project or life context and connects\nreports to the processes they describe. As new evidence arrives, it returns to\nthe earlier interpretations and decisions that evidence may change. In writing\nfeedback, the assistant reads the whole before diagnosing a passage, distinguishes\nwhat the text says from an interpretation or artistic alternative, and follows\nthe consequences of the human's chosen revisions. Both workflows use the same\nmethod: the assistant starts with a coarse account, develops the parts that\nmatter, and reconsiders the whole in light of what it finds.\n\nMemory and feedback have their own `life_modeling_context` purposes:\n`agent_memory` for the agent's own work, `user_memory` for what a user reports,\nand `human_author_feedback` for writing feedback. For memory that survives a\nrestart, configure a private `LIFE_SIM_STATE_FILE`; use one server process per\ndatabase. A correction keeps the earlier account and is distinct from a real\nchange over time. Dependent interpretations need review; they do not update\nautomatically. Transcript capture is optional and off by default;\n`life_memory_transcript_configure` turns it on for one memory context. It stores\nonly the visible messages that the caller or chat integration supplies, separately\nfrom the process model and its interpretations. The server never observes a\nconversation itself.\n\n## The construction record\n\nThe model and its Understanding Graph are the modeler's understanding, not a report\nabout it: what is done and not recorded cannot be picked up by the next agent. Every\nEvent that carries a Cut has a description of what happens in it, so its numbers mean\nsomething. Choices, ideas, predictions and voice decisions are recorded as\nUnderstanding Nodes linked to the events, Cuts and passages they concern, and outside\nreviews are recorded under their actual reviewers. `life_model_outline` shows the\npresent state with its notes at a chosen depth, and `life_construction_replay` replays\nthe whole development step by step, each note beside the records as they were when it\nwas written. An agent continuing unfamiliar work recovers its history through\nreplay and an outline before changing it. When the exact previously read revision\nand context are retained, the agent reads subsequent changes and the records\nrelevant to the current task.\n`life_construction_export` and `life_construction_import` carry the whole history to\nanother engine with the same hashes, so the worked examples can be replayed anywhere.\n\n## Model what your application needs\n\nChoose your own processes, categories, relationships, and depth. Begin with\none process, use a supplied starter, or author a model directly. A complete\nperson model, the Book's vocabulary, and the optional semantic, narrative,\nand decision layers are not prerequisites for every application.\n\nThe AI modeler can use this external model to connect evidence, compare\nexplanations, explore possibilities, and guide further inquiry. It also chooses\nwhen to replace or refine its categories. The engine checks supported structure\nand preserves explicit revisions; it does not choose a universal psychology.\nSome templates load structure, while Story and Decision compile particular\nexperimental behavioural assumptions. Choose them deliberately. See the\n[application guide and revision example](https://github.com/emergent-wisdom/meaning-model/blob/main/docs/examples/APPLICATION-CATEGORIES.md).\n\n## General-purpose modeling and optional Jev estimation\n\nStart with `life_general_modeling_start`. The general workflow supports\ndomain-defined processes across technology, adoption, institutions, markets,\nphysical systems and other applications. `life_world_model_build` constructs\nthe initial model and graph from a compact scaffold. Its required context review\nstarts with the enclosing system and longer-term developments before focal\nprocesses; represented context, unknowns and justified exclusions become\nUnderstanding Nodes. It also requires consideration of authored numerical\njudgments, useful native conceptual decomposition, and meanings across dates or\nperspectives. The agent may explain why a dimension is unnecessary or unresolved;\nthe tool does not demand arbitrary scores or depth. These checks do not prove\ncompleteness or causal relevance.\nThe builder's preview can evaluate Jev\n`initialEstimate` questions; applying that exact proposal adopts the initial\nvalues as estimates without another provider call. `life_process_estimate`\nuses bounded questions to create typed process-value proposals;\n`life_process_estimation_record` saves the exact proposal, process records and\nreview in the graph. Estimates retain their status and evidence, and do not\nsilently become observations in accepted runtime history.\n\nSet `MEANING_MODEL_ESTIMATOR=typesafe` and `TYPESAFE_API_KEY` to use Jev for\nbatched structured judgments. This is independent of the storytelling add-on\nand is off by default. The LLM chooses the scope and questions; tool code\nhandles repetitive record construction. Reduced end-to-end cost and latency\nrequire measurement, including setup and review work.\n\nRead `life-sim://guide/general-modeling` for the complete workflow and limits.\n\n## Install\n\nInstall the package and matching engine together. Prebuilt installation needs\nthe matching published GitHub release assets; a candidate without those assets\ncan use the explicit source build below.\n\nIn a directory where you want to keep the installation, run:\n\n```sh\nnpm install @emergent-wisdom/meaning-model-mcp@0.6.8\nnpx meaning-model-mcp --install-engine\n```\n\nNode.js 22.18 or later is required. The explicit `--install-engine` command\ndownloads the executable for this package's exact version from the official\n[`emergent-wisdom/meaning-model` GitHub release](https://github.com/emergent-wisdom/meaning-model/releases).\nIt supports macOS 14+ on Apple Silicon or Intel, Linux x64 with glibc 2.35+\n(for example Ubuntu 22.04+), and Windows x64. It needs no Rust compiler.\nOther platforms can use the source build or an existing compatible engine.\n\nThe installer checks the exact release tag, target, size and SHA-256 digest\nreported by GitHub's release API before replacing the package's default\n`rust-engine/target/release/life-sim-engine` executable (`.exe` on Windows).\nDownloads are bounded and use GitHub HTTPS URLs; replacement is atomic, and a\nfailed download or checksum check preserves the previous executable. The\ninstaller never writes to the `LIFE_SIM_ENGINE_BIN` override. Close a running\nengine before reinstalling on Windows, where a process can lock its executable.\n\nThis verifies integrity against the official GitHub release channel. It trusts\nGitHub and the repository's release maintainers; it is not independent author\nsigning, OS code signing, notarization, or proof of reproducible builds. A SHA-256\nfile is also published with each executable for manual checking. The automated\ninstaller requires GitHub's own asset digest instead of trusting only that file.\n\nnpm installation and ordinary MCP startup never download or build an engine.\nRun `--install-engine` again after installing a new package version. If the\nmatching release is unavailable, the installer fails explicitly. To build the\nincluded source instead, run:\n\n```sh\nnpx meaning-model-mcp --build-engine\n```\n\nThe source build requires Cargo, a C compiler, and native build tools. It may\nfetch dependencies pinned by `Cargo.lock`; Rust dependencies include bundled\nSQLite. Alternatively, set `LIFE_SIM_ENGINE_BIN` to a compatible engine already\nbuilt for this machine.\nFor a release-candidate test, replace the package name with the path to the\nreviewed `.tgz`; use `--build-engine` until that candidate's release assets exist.\n\n## Connect\n\nAfter installing or building the engine, choose a private data directory outside\n`node_modules` and create that directory. Replace both example paths below with\nabsolute paths on your computer, then point an MCP client at the installed launcher:\n\n```json\n{\n  \"mcpServers\": {\n    \"meaning-model\": {\n      \"command\": \"node\",\n      \"args\": [\"/absolute/path/to/node_modules/@emergent-wisdom/meaning-model-mcp/mcp-server/bin/meaning-model-mcp.mjs\"],\n      \"env\": {\n        \"LIFE_SIM_STATE_FILE\": \"/absolute/private/path/meaning-model.sqlite\"\n      }\n    }\n  }\n}\n```\n\nThe default command serves only MCP on stdio. Run `meaning-model-mcp --help`\nfor setup options. Existing `life_*` tools, `life-sim://` resource URIs,\nserialized schemas, `LIFE_SIM_ENGINE_BIN`, and the `life-sim-engine` binary\nretain their compatibility names.\n\n### Keep and continue your work\n\n**Keep `LIFE_SIM_STATE_FILE` set to the same absolute path across sessions and\npackage upgrades.** The engine creates the SQLite database on the first successful\nwrite, but its parent directory must already exist. Run only one MCP server\nprocess per database; stop the previous process before another client uses that\nfile. Without a state file, models, stories and memory are lost when the process ends.\nAsk the assistant to check `life_engine_status` to confirm persistence is enabled.\n\nThe database retains Rust-owned models and their revisions, narrative and\nUnderstanding Graphs and their recorded history, worlds, and candidates. MCP\nworld handles/index, request receipts, annotations, pending writer plans, and\nestimation exchange records remain process-local and are not restored on restart.\nSave authoring decisions and results in the graph to retain them; a pending tool\nproposal or chat conversation alone is not a saved construction record.\n\nThe 32-entry Node model-summary cache does not limit saved history to 32\nrevisions. Rust enforces storage limits, currently 512 model revisions and\n256 MiB of model definitions; `life_engine_status.engine.nativeLimits` reports the\nrunning engine's limits.\n\nTo continue, ask **“Find my saved work and continue this story.”** The assistant\nshould:\n\n1. Call `life_saved_work_list` with the same `accessScopes` used for the work, page\n   through the results, and select the intended exact graph revision. Alternative\n   branch heads remain separate choices. The list covers graph-backed work only;\n   it does not enumerate models without a graph or work hidden by those scopes.\n2. Use `life_construction_replay` and `life_model_outline` to recover unfamiliar\n   work. When the exact prior revision and context are retained, read subsequent\n   changes and the records relevant now before deepening or reviewing it. Reuse those\n   scopes in subsequent calls; an empty or different scope may hide private work.\n3. Develop and review through the MCP tools, saving model revisions, story text,\n   and reasoning in the graph. With the storytelling add-on, `life_story_deepen`\n   helps choose useful depth, and scene review checks the proposed continuation.\n4. Use `life_model_viewer_open` on the resulting revision to read and inspect it.\n   The local viewer link ends with the MCP process; opening it does not publish.\n5. Ask for a portable backup with `life_construction_export`, using\n   `destinationPath` for an absolute private JSON path outside the installation,\n   such as `/absolute/private/path/backups/story-2026-09-27.json`. Create its\n   parent directory first and choose a new filename: exports never overwrite.\n   For books, prefer this file route; the assistant gets a compact receipt rather\n   than the whole history in chat. To restore it, use `life_construction_import`\n   with `sourcePath` and `requestId` on the destination engine.\n\nPortable files contain the selected model-bound graph's retained model and graph\nhistory, up to 256 MiB. Export requires scopes covering the complete retained\ngraph history. Import checks the bundle checksum before registering revisions;\nit leaves the source file unchanged. It is a sequence of registrations, so a\nlater failure may leave earlier imported revisions in the destination. Worlds,\ncandidates, process-local MCP work, and other models mentioned only through\nexternal references are not included. Manuscript files are exports, not a\nsecond authoring record to edit separately. Construction history includes old\ndrafts and reviews: keep it as a private backup or transfer, not a publication\nartifact. A published manuscript comes from `life_narrative_render`, with only\ndeliberately reviewed model or companion material selected for publication.\nAdvanced callers can omit `destinationPath` for inline export, limited to 1 MiB,\nand supply that result as `history` on import instead of `sourcePath`.\n\n### Optional storytelling add-on\n\nThe authoring workflow below describes delegated story generation. For a human\nwriting their own work, use [human-author feedback](https://github.com/emergent-wisdom/meaning-model/blob/main/docs/HUMAN_AUTHOR_FEEDBACK.md)\ninstead; feedback alone does not create a project or rewrite the manuscript.\n\nScenes can contain ordered, independently editable passage nodes. The shared\n`life_narrative_edit` tool supports split, merge, move, reorder, and exact text\nreplacement while preserving earlier graph revisions and the frozen source.\nIt reports affected reviews for reassessment and is available without the\nstorytelling add-on as well.\n\nThe package includes opt-in structure exploration, overall character life\ntrends, model-depth review, scene preparation, review, and commitment, plus advisory chapter or\nsection purpose review. To enable\nthem, add `MEANING_MODEL_ADDONS` to the server configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"meaning-model\": {\n      \"command\": \"node\",\n      \"args\": [\"/absolute/path/to/node_modules/@emergent-wisdom/meaning-model-mcp/mcp-server/bin/meaning-model-mcp.mjs\"],\n      \"env\": {\n        \"LIFE_SIM_STATE_FILE\": \"/absolute/private/path/meaning-model.sqlite\",\n        \"MEANING_MODEL_ADDONS\": \"storytelling\"\n      }\n    }\n  }\n}\n```\n\nUse `\"storytelling,alien\"` to enable the alien ideation add-on as well, or\n`\"alien\"` alone. Keep the same `LIFE_SIM_STATE_FILE` and any existing\n`LIFE_SIM_ENGINE_BIN` in the `env` object, and restart the MCP server. To let\nTypeSafe's Jev score the estimator tools\n(`life_estimate_cut_shares`, `life_model_ingest`, `life_narrative_alignment_audit`) instead of\nreturning their questions to you, add `MEANING_MODEL_ESTIMATOR: \"typesafe\"` and\n`TYPESAFE_API_KEY` to the same `env`; with that setting the text given to those\ntools is sent to TypeSafe. Everything else stays local. The add-on exposes twelve tools: `life_story_structure_explore`,\n`life_story_trajectory_explore`, `life_story_trajectory_revise`,\n`life_story_author_record`, `life_story_life_trends`,\n`life_story_model_depth_review`, `life_story_model_depth_record`,\n`life_story_scene_prepare`, `life_story_scene_review`,\n`life_story_scene_commit`, `life_story_purpose_review`, and `life_story_deepen`, plus the\n`life-sim://addon/storytelling` resource and the `life_story_scene_start`,\n`life_story_structure_explore`, `life_story_purpose_review`, and `life_story_deepen` prompts.\nIt uses the existing engine and narrative store. No extra package is needed.\n\nThe narrative graph is the authoritative authoring record. Create the model\nand graph before developing story material; store candidates, seed draws and\nalternatives, drafts, assessments, selections, local revisions, and disclosure\nplans through the tool. Files and PDFs are exports of graph content, not a\nparallel manuscript or model. Use `life_story_author_record` for authoring\nmaterial and concise Understanding Nodes. Numerical exploration and revision\npersist their results directly; neither accepts them as world facts.\n\nFor new work the LLM develops or reuses an author model, using supported real\nmaterial or an explicitly invented persona within the human's delegation.\nThe author at composition time, their writing history and reasons for this\nwork inform concrete writing choices. The author, narrator and characters\nremain distinct. After drafts and substantive revisions, the LLM assesses the\nauthor's voice and each relevant principal character using actual passages\nand model process evidence. It records intended versus observed effects,\nuncertainty, and the smallest repair or a reason to keep the text through\n`life_story_author_record`, `kind: \"assessment\"`, as actual Understanding Nodes\nwith reviewed hashes and evidence links. An external review alone is\ninsufficient; literary effectiveness remains advisory.\n\n`life_story_deepen` prepares a later revision round of an existing work. It\nbinds the baseline graph, model, rendered text and selected author model, with\ndepth and purpose-review tasks. Default `revisionScope: \"local\"` preserves\npremise, cast and ending; `structural` permits justified larger changes within\nthe agreed brief. The LLM saves before-analysis, a revision plan and\nafter-analysis as Understanding Nodes and uses existing tools to revise.\nKeeping successful work is valid; there is no reroll or length quota. The\ntool and same-named prompt are read-only and do not certify improvement.\n\nFor an existing book whose lives and causes are already modeled, the assistant\ncan set `lifeTrendsNodeId: null` and supply `modelEvidenceRefs` to\n`life_story_deepen` or `life_story_model_depth_review`. References such as\n`event:<id>` or `process:<id>` select actual records from the bound model;\nthey do not certify sufficient life coverage. Inspect the evidence, record\nunknowns, and refine only what the selected passage needs. This mode supports\nreviewing and revising existing work; new-scene commitment still requires its\nlife-trends dossier and matching depth assessment.\n\nUse `life_narrative_edit` to replace or subdivide existing passages while keeping\ntheir earlier revisions. Inspect the retained links and review affected prose.\nScene commitment appends new material; an append-based replacement requires\nexplicitly retiring and rewiring the superseded passage. Review the exact final\nrendered topology for duplicates, gaps and changed order, preserve the baseline,\nand never export an intermediate gap as the finished revision.\n\n\nThe calling LLM automatically constructs or reuses the principal cast's overall\nlife trends during exploration and before preparing scenes for commitment; the user need not request the step or\nfill in a dossier. Store a new dossier with `life_story_life_trends` when\nneeded. It validates and stores authored coarse life\nphases, trends across those phases, and explanations for their\ndevelopments. The scene workflow requires this dossier and connections from\neach present or affected principal character to relevant trends; review findings address\ntheir continuity with the prose. Missing dossiers and incomplete structures\nare rejected. The LLM must still create a substantive model and interpret the\ndraft. A brief character sketch or the immediate crisis is not the intended\nscope, and an undecided later life can remain open. Exploratory drafts and\ncandidates can precede a resolved depth review. Follow the questions they reveal\nback into the model, open further processes and relationships, and repeat.\n\nBefore prose and after consequential model, trajectory, causal, or disclosure\nrevisions, the LLM automatically reviews whether the model explains the\nstory's consequential choices and outcomes. `life_story_model_depth_review`\nreads the actual bound model and stored focus, life dossier, and selected\ncontext. `life_story_model_depth_record` saves a coverage explanation and\nfindings as a scoped Understanding Node, with validated references to graph\nnodes or model JSON Pointers. Relevant lives and flaws, concepts, constraints,\ninstitutions, causes, and disclosure processes can be examined without a\nfixed taxonomy or depth quota. Preserve adequate detail; repair the smallest\nexplanatory gap and reassess.\n\nSupply `modelEvidenceRefs` alongside the dossier to read the actual records\nbehind the chosen focus. The packet also resolves model anchors from selected\ngraph context, preserving full records with their questions and units. These\nremain available when the whole model is too large to include. Omitted records\nand unavailable frozen-runtime evidence are identified explicitly; follow the\nrelevant dependencies before calling an account sufficient. The bundled\n`life-sim://example/book-of-conditions-modeling` shows this method without a\ncompulsory category list.\n\nScene preparation requires `modelDepthReviewNodeId` for a current\nassessment covering its source, life dossier, and context. Changed evidence\nrequires reassessment; unrelated author notes or drafts do not. Missing or\nstale records fail preparation; unresolved findings produce a packet with a\ncommitment blocker. Assessments and drafts can be saved while gaps are\nresolved. This is a check on\nthe authored explanation, not a literary-quality score. The model read is\nadministrative, not filtered by narrative scopes, and its initial values are\nnot current world values. The LLM must identify missing evidence honestly.\n\nWhen rebinding the narrative graph after a model revision, retain old depth\nassessments as historical records but remove their predecessor-model anchor\nedges from the successor through `life_narrative_revise`. Prior immutable\ngraphs retain the exact evidence. Never retarget old findings to new values;\nrecord a fresh review with new model anchors before affected scene commitment.\n\nNumerical trajectory exploration is optional when a declared quantitative\nquestion benefits from sampling; a new life or Event needs no sampled scores.\nFear/love is likewise an available lens, selected explicitly or retained from\nexisting authored readings, rather than an automatic backlog for every new story.\nNumerical exploration samples actual candidate values at selected\nevent or whole-life points, including emotional dimensions. The LLM defines\naxes with explicit meanings, comparison questions, units, and bounds, then\nsupplies expected baselines and fixed values. Optional disjoint allocation\ngroups retain their declared totals. `randomness` (0–1, default 0.5) controls\nvariation; `candidateCount` (default 3, maximum 8) separately controls how many\npossibilities to explore. A seed enables replay.\n\nThe LLM reviews coherence and storytelling potential and can repair a promising\ncandidate with `life_story_trajectory_revise` rather than discard the character.\nLocal revisions bind the parent hash and reasons, preserve unlisted and fixed\nvalues, and check bounds and allocation totals. Retain original samples and\nrevision history. Category or point-time changes require an explicit new\nproposal. Character dossiers can retain `trajectoryProposal: {definition,\ncandidate}` with `trajectoryRecordNodeId` as numerical evidence, with prose\nsummaries explaining the values. Storage verifies the exact candidate against\nthat graph record and preserves its access scopes.\nCandidates are unaccepted creative hypotheses, not physical simulations or\ncalibrated psychological measures.\n\nRecord concise assessments and revision reasons as Understanding Nodes through\n`life_story_author_record`, under its named author-understanding root. Use\n`externalized_reflection`, a holder, author scopes, excluded rendering, and\nspecific `about`, `refines`, or `supports` links to the relevant evidence.\nThese are authored explanations, not hidden model reasoning or accepted world\nfacts. Numerical revisions persist their reasons automatically; the LLM records its\nevaluations separately. The tools do not make literary judgments.\n\nStructure exploration draws from a bounded bank of common English words, or\nuses a caller-supplied word, to help the calling LLM propose possible events,\ncharacters, relationships, storylines, or names. For new principal-character,\nplace, or organization names, the LLM automatically uses `targetKind: \"name\"`\nand omits `seedWord` to draw a random inspiration. Supply culture, language,\ngenre, tone, and existing names as context; use sound, rhythm, or associations\nto develop names that fit. Preserve established names unless a change is\nrequested, and distinguish invented etymologies from real linguistic facts.\nThe bank remains 160 reviewed words, not the STLM vocabulary. Record the seed task and proposed alternatives with `life_story_author_record`;\nproposals remain unaccepted.\n\nThe calling LLM automatically performs purpose review after completed chapters,\nsignificant turning points, completed parts or works, and consequential\nrevisions. A scene commit's `nextStep` reminder is conditional: the LLM decides\nwhether the relevant unit has finished. Supply context for life trajectories,\nexpectations, causes, aftermath, and authored disclosure. Review remains\nqualitative and advisory; keeping the text unchanged is valid, and missing\ncontext may leave the result unclear. There is no rewrite quota or review\ntimer. Random-word exploration and purpose review return tasks for the calling\nLLM rather than invoking another model. Numerical exploration persists actual\nsamples for that LLM to interpret. Depth-recording persists the LLM's findings;\ndepth preparation does not invoke another model or produce its own verdict.\n\nOptional anticipation, focal change, and adaptation Events already exist in\nthe core `change_arc_scaffold`. Their use does not impose a three-beat story:\nshock can be welcome or expected, and adaptation can begin beforehand, overlap,\nor remain incomplete. Model character experience separately from author\nprocesses for intended reader anticipation and surprise.\n\nWith the variable unset, the server's default surface is unchanged. The shared\nmodel and laws remain general-purpose, and existing low-level tools remain\navailable. See the [storytelling add-on guide](https://github.com/emergent-wisdom/meaning-model/blob/main/profiles/STORYTELLING_ADDON.md)\nfor the workflow and the distinction between mechanical checks and authored\nprose-review findings.\n\n### Optional alien add-on\n\nSet `MEANING_MODEL_ADDONS=alien` (or `storytelling,alien`) to add world-diversity\nideation from *Ontology of the Alien*. The server writes each role's task with\nonly what that role may see:\n\n- a target-blind builder turns a seed word into an invented world;\n- a purpose-blind solver solves the problem inside that world;\n- a compiler brings the operative mechanism back into the problem's domain.\n\nThe add-on guide and generated role tasks supply the search procedure; its\nbundled paper is an optional reference. Every cell of the paper's condition\nmatrix can be run, from direct proposals with a Semantic Tabu archive or the\ncurated map to map-conditioned compilation.\n\nCurators keep three revisable ontologies: mechanism families, claimed outcomes\nand causal world regimes. A new family is admitted only when the recorded\nequivalence test says the primary causal operator changed. Diagnosis of those\nontologies informs which world to commission next, including family and outcome\ncombinations no candidate has yet; the choice stays with the caller. Promising mechanisms are transferred onto a\ntarget model, with every role mapped and every disanalogy stated.\n\nA second judge can check each curator decision: with the Jev estimator\nconfigured it answers the same comparison without seeing the curator's reasons,\nand disagreements appear in the diagnosis. A candidate between families can carry\na graded membership, shares with a remainder, beside its category. Target-blind\nworlds can be exported as a content-addressed library and imported into another\nsearch, which then starts at the solver.\n\nEverything lives in the narrative graph as Understanding Nodes. Every task text\nis stored, so each output cites what its role actually saw. Graded fits and\nweighted selections follow the Meaning Model's Cut rule. Each ontology can be\nexported as Meaning Model concepts and specialization relations. Worlds are\ntextual thought experiments and transfers are ideas, not evidence. See the\n[alien add-on guide](https://github.com/emergent-wisdom/meaning-model/blob/main/profiles/ALIEN_ADDON.md). It exposes nine tools:\n\n- `life_alien_search_start`\n- `life_alien_task`\n- `life_alien_record`\n- `life_alien_ontology_revise`\n- `life_alien_decision_check`, the second judge\n- `life_alien_search_diagnose`\n- `life_alien_atlas`\n- `life_alien_worlds_export`\n- `life_alien_worlds_import`\n\nIt also adds the `life-sim://addon/alien` guide, the bundled paper as\n`life-sim://theory/ontology-of-the-alien`, and the `life_alien_start` prompt.\n\n### Registry clients using npx\n\nThe official MCP Registry name is `io.github.emergent-wisdom/meaning-model`.\nIts `npx` configuration requires `LIFE_SIM_ENGINE_BIN`: an absolute path to\nthe engine installed or built above. This avoids assuming that the package\ninstance in an `npx` cache is the same installation you prepared. Registration\ndoes not download or build the engine for you.\n\nFrom your installation directory, print the engine path:\n\n```sh\nnode -p 'require(\"node:path\").resolve(\"node_modules/@emergent-wisdom/meaning-model-mcp/rust-engine/target/release\", process.platform === \"win32\" ? \"life-sim-engine.exe\" : \"life-sim-engine\")'\n```\n\nUse that path when a Registry client asks for `LIFE_SIM_ENGINE_BIN`. Also set\n`LIFE_SIM_STATE_FILE` in that client's server environment to your existing private\ndatabase path, even if the client only prompts for the engine path. A manual\nconfiguration looks like this:\n\n```json\n{\n  \"mcpServers\": {\n    \"meaning-model\": {\n      \"command\": \"npx\",\n      \"args\": [\"--yes\", \"@emergent-wisdom/meaning-model-mcp@0.6.8\"],\n      \"env\": {\n        \"LIFE_SIM_ENGINE_BIN\": \"/absolute/path/to/life-sim-engine\",\n        \"LIFE_SIM_STATE_FILE\": \"/absolute/private/path/meaning-model.sqlite\"\n      }\n    }\n  }\n}\n```\n\nThe override is not required when directly starting the same installed\nlauncher whose engine you installed, as in the first configuration. Install or\nbuild the matching engine when upgrading the package. On Windows its filename\nends in `.exe`.\n\nThe [complete server guide](https://github.com/emergent-wisdom/meaning-model/blob/main/mcp-server/README.md) describes modeling, the\noperational guidance, tools, quotas, persistence limits, and the boundary between\nimplemented mechanisms and research proposals. This remains experimental\nsoftware; a released package does not establish the papers' learning claims.\n\n## Prepare platform engines\n\nThe source repository includes `.github/workflows/engine-release.yml`. It builds\nfrom an existing `v<package-version>` tag using a pinned Rust toolchain and\n`Cargo.lock`. GitHub Actions run native builds on macOS arm64, macOS x64, Linux\nx64 and Windows x64. Each job checks packaging and offline installation behavior,\nstarts the real Rust engine, then checks an MCP connection and engine status.\nOnly passing jobs upload the version-named executable and its `.sha256` file.\n\nOnce the reviewed source, workflow and matching tag are pushed, select **Build\nengine release** in the repository's Actions tab. Run it with `tag: v0.6.8` and\nleave `create_draft` false for a build and smoke run that only uploads workflow\nartifacts. Set it true to create a draft release after all four platforms pass.\nPushing a new `v*` tag also runs the workflow and prepares a draft release.\nExisting releases are never overwritten by the workflow. Only its final draft\njob receives `contents: write`; build jobs have read access and action revisions\nare pinned to full commit IDs.\n\nReview the four platform jobs, all eight release assets and checksums. Publish\nthe reviewed GitHub draft before publishing the matching npm package. Drafts\nare unavailable to the installer. Then install the exact candidate tarball and\nrun its `--install-engine` command on each supported platform to check public\ndownload and execution. A local macOS check cannot establish Linux or Windows\ncompatibility; the four jobs and public-download checks are release gates.\n\nThe **Verify public engine release** workflow runs those real public\ndownloads and Rust/MCP checks on all four platforms. It runs when a release is\npublished or can be dispatched with the exact release tag, and never publishes\nor modifies release assets.\n\nThe release assets are raw executables named\n`life-sim-engine-v<version>-<Rust target>` (`.exe` on Windows), plus an adjacent\n`.sha256` file. They contain no archive paths or extraction step. The package\nand Rust engine versions, Git tag, and MCP metadata must agree. Review OS code\nsigning separately if you need it; this workflow does not sign or notarize.\n\n## Prepare and publish the npm release\n\nAfter the repository checks pass, create the complete tarball from the repository\nroot:\n\n```sh\nnpm --prefix mcp-server run pack:release\n```\n\nUse the exact `tarball` path reported by that command. Direct packing or\npublication of the `mcp-server` source directory is blocked because it omits\nthe Rust engine and required reading resources. Install that tarball and test\nthe explicit engine installation/build and MCP connection before approving it\nfor release.\n\nRecord the reviewed tarball's checksum and inspect the publication preview:\n\n```sh\nrelease_tarball=\"/absolute/path/to/emergent-wisdom-meaning-model-mcp-0.6.8.tgz\"\nshasum -a 256 \"$release_tarball\"\nnpm publish \"$release_tarball\" --dry-run --access public --ignore-scripts --registry=https://registry.npmjs.org/\n```\n\nPublication remains a separate release-owner decision. After approval of that\nexact tarball, authenticate with an npm account that can publish to\n`@emergent-wisdom` (`npm login` if needed), verify its identity, then publish:\n\n```sh\nnpm whoami --registry=https://registry.npmjs.org/\nnpm publish \"$release_tarball\" --access public --ignore-scripts --registry=https://registry.npmjs.org/\nnpm view @emergent-wisdom/meaning-model-mcp@0.6.8 version dist.integrity --registry=https://registry.npmjs.org/\n```\n\nA dry run does not establish registry authentication or scope access. Any change\nto the package requires a new tarball and review. See the\n[npm publication documentation](https://docs.npmjs.com/cli/v11/commands/npm-publish/)\nfor tarball publication and registry behavior.\n\nPublish the official MCP Registry metadata only after the referenced npm\nversion is available. The root `server.json` name must match the package's\n`mcpName`; both version fields must match the reviewed release. With release\napproval, run the following from the repository root:\n\n```sh\nmcp-publisher login github\nmcp-publisher publish server.json\n```\n\nVerify the entry through the\n[official Registry API](https://registry.modelcontextprotocol.io/v0.1/servers?search=io.github.emergent-wisdom/meaning-model).\nPublisher credentials are local authentication material, never release files.\n\n## Licensing\n\nCode is MIT licensed; authored papers and documentation are CC BY 4.0.\nSee [LICENSE](https://github.com/emergent-wisdom/meaning-model/blob/main/LICENSE), [LICENSE-CONTENT](https://github.com/emergent-wisdom/meaning-model/blob/main/LICENSE-CONTENT), and [NOTICE](https://github.com/emergent-wisdom/meaning-model/blob/main/NOTICE)\nfor scope, attribution, and third-party material.\n","readmeFilename":"README.md"}