{"_id":"memorykit-mcp-server","_rev":"11-714d8401b912fb3615638dbed0e03889","name":"memorykit-mcp-server","dist-tags":{"latest":"2.0.0"},"versions":{"1.0.0":{"name":"memorykit-mcp-server","version":"1.0.0","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.0.0","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"c4c76c5eeaefd8adec00de8b1bf3b43470027c1e","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.0.0.tgz","fileCount":97,"integrity":"sha512-ajXGGRMk6d9M73jQkH6ISCXpbYXDqOkoKD0q9zon45ECZQO82YXK0fSKkNyMOxkZcubWyZA4Vdo9wu0ZuJ1fCQ==","signatures":[{"sig":"MEUCIQDOByM635GnlfZ7iiNcAuxOo3QTmpR32RWX0Qfnc0E3UwIgWYy1Oam3Qa7Y1ex1HWxPz6kglLxuEodU1gTceBSaIn0=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":355248},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"351f1e8a4040aad87c59a36137b81d8a7b7bd73e","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"10.8.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"20.20.2","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.0.0_1781794738914_0.6607537222211446","host":"s3://npm-registry-packages-npm-production"}},"1.0.1":{"name":"memorykit-mcp-server","version":"1.0.1","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.0.1","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"7eabc8e39f6b88a713d549989d4871163da4811f","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.0.1.tgz","fileCount":97,"integrity":"sha512-PaPBGOKK29f2Jvm1UrL7EWcuI+oLmvdKMIc4FClq8cgrpJDiijNys5P0rI4FF4eqzzcEYG+l3c/wYQPsB3NU8Q==","signatures":[{"sig":"MEQCIEXGhfolI49+WGb4TUdoxfsknSKG0ZyBgta/Cgk7H23TAiBhzZMsQHu0r2hBbu9Up4YCOUknOmhL+blt2PW479i19Q==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":357676},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"c11ab9a137a3067eb6a8847ea5331bcc9cddf972","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"10.8.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"20.20.2","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.0.1_1781828194042_0.7890035136824276","host":"s3://npm-registry-packages-npm-production"}},"1.1.0":{"name":"memorykit-mcp-server","version":"1.1.0","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.1.0","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"21dca6a3256ae11ee03340cf5de4d6ebde53f64f","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.1.0.tgz","fileCount":99,"integrity":"sha512-FlzFR3X2kdRLmZBWqU3BrN4ZOCrX+ttCTLsXeFvyxQy7rKGKwVCGmklQdhqkgo41pSeoA2j4dulqDIRrmrSvCw==","signatures":[{"sig":"MEUCIFcuwYdl8efhBDMeNnZGZzn5bJfcxvGOKaqhBW1Z6KIgAiEAhH7EW+pk2YQFdsiovwhXC8r0x8lhv3q6rCsRiHFE/+A=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":371775},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"471c3abed8db04f9ee34a050d0be7aecd084d469","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"10.8.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"20.20.2","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.1.0_1784045117265_0.5748862134847901","host":"s3://npm-registry-packages-npm-production"}},"1.1.1":{"name":"memorykit-mcp-server","version":"1.1.1","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.1.1","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"9674071df5d41283206ad9fbc877d5f3123c6d87","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.1.1.tgz","fileCount":99,"integrity":"sha512-AYZGsNw5VohUhCibG+8fQaGh/RGUyYk9qzzw7CbYCtXT7BIZdE7SBWLLSTKHIKWbJfYRsfh7d4yvmkpkHysvRg==","signatures":[{"sig":"MEQCIBuaGGuPY9J6sNQdCZEV+sRkbglH2bL0PJAANbOmr3J4AiBQD2UsE1ymRv4EdoozQaqeH0hzNCAWKEA9NbYZb5nu7g==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":375356},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"c574aaf05a2441157e0546dba7ff3b17399c9166","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"10.8.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"20.20.2","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.1.1_1784562854261_0.9741503016382433","host":"s3://npm-registry-packages-npm-production"}},"1.2.0":{"name":"memorykit-mcp-server","version":"1.2.0","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.2.0","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"f3691fb7bb358153c881688eb6be15eb812f73e6","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.2.0.tgz","fileCount":99,"integrity":"sha512-JhJHmBiS+8fBcKuwQOPMYH4bgUL3AcipTCAN3DzPN2LP0Sq/KWlx3ahZ1PL+FArOpUvpI8JmeSvjQ5VoSYjKGw==","signatures":[{"sig":"MEYCIQDrSD3hQjgZqKHFpIAd9fHMMH73vUIjFFdyC0jur5DyOQIhANYmIx8z7WVF1hoTYvFLd1dfa38Rm97FmK9Ogc2P8WPb","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":387526},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"f1e26d51e2b82a143b386b0dc259aae2ef9ac725","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"10.8.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"20.20.2","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.2.0_1784636644419_0.5554139359880923","host":"s3://npm-registry-packages-npm-production"}},"1.3.0":{"name":"memorykit-mcp-server","version":"1.3.0","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.3.0","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"45144c3b0ee0de1538879ac151d817a6441db0ff","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.3.0.tgz","fileCount":99,"integrity":"sha512-H2si3LHi/lc/bZHyXA8xvJeE+o2HOFPZr6OunoTqqfyor0n+SVQaZZUi6kVWWOaju1ytRWF6+41dpfy/1gtbBQ==","signatures":[{"sig":"MEUCIQC4k4I7Bd/ekGc7ZbBcxQhd+lvHS05PeIkybCPRDtyqOAIgQxkzczGd2091ZsRECJ1xzpka5Eus0CBxJfZOAqTkSIs=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":402940},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"7501b1503f5156280b5b4791e3cd44efa6c1a86c","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"10.8.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"20.20.2","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.3.0_1784712888380_0.3999806217360382","host":"s3://npm-registry-packages-npm-production"}},"1.3.1":{"name":"memorykit-mcp-server","version":"1.3.1","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.3.1","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"716352b2a6cf417dd37d6bbc1df306a5de3daed3","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.3.1.tgz","fileCount":99,"integrity":"sha512-5tbbs/3bjY6Cf3skZkZkdyqwg/pCbBD5GFzIXGeIlefFUkKYa3FgduQxMn1sJX55lVUQOcv7Q7VTGeiw9GZYKw==","signatures":[{"sig":"MEUCICLh4UQK1kTFjprhQI68YF75cXwuK62oql1CK+67yoYCAiEA5vLSMRMQgkoDd92Mm4erpySo0T0wT2qCl0uSO+5/EKA=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":404268},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"490057be9221b6781de1b668a9d755d6761d72c6","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"10.8.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"20.20.2","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.3.1_1784728122937_0.3262482450094024","host":"s3://npm-registry-packages-npm-production"}},"1.4.0":{"name":"memorykit-mcp-server","version":"1.4.0","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.4.0","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"358631bb06a6df5f127ad24f12de419f71fc8d3d","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.4.0.tgz","fileCount":113,"integrity":"sha512-eDZB4c8cZ78WPTRDzvmeCa2krXdN6M8MaFWP1/hShq2gdxFT8Zf1II0hgZMtc6ltrHDZuciHjy1GRClWByCfAA==","signatures":[{"sig":"MEUCIBEsxBIu+3FbL7IVVBhiZoa0Mll9CX+qguJFijvbJfdgAiEAg78HLEZ3/11AxFhv8X1BfK98b/zSVDaWuGOVtxSsjbY=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":453481},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"16feca8b152b92457838162fbe961e7334ca53cc","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.6.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"24.12.0","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.4.0_1785417830363_0.37608632249919305","host":"s3://npm-registry-packages-npm-production"}},"1.4.1":{"name":"memorykit-mcp-server","version":"1.4.1","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.4.1","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"b2614e88bc320734e3c9221fa225c67fba580b0e","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.4.1.tgz","fileCount":113,"integrity":"sha512-9ybnie5QN1IM9EfyelwfoXK55BVs3+baqUzyQSPNJC3WQKtNYWcLF9LFTTbMFqQCzgQMemvEmAwPf7Dz+qvCXA==","signatures":[{"sig":"MEUCIEev+2/Mc2EtKPH0aiCpFGEbUhATPTwx2zGOTiXgPG6HAiEAxjI3cVpWIMYNM4/sRFzPKXgLGMSR7QWZyzyxXN9znUA=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":453816},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"7b52d512dfb32eb605782085960adb4b7bfb429e","mcpName":"io.github.rapozoantonio/memorykit","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.6.2","description":"Cognitive memory for AI coding assistants — persistent memory across conversations for Claude, Copilot, and Cursor","directories":{},"_nodeVersion":"24.12.0","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.4.1_1785421998847_0.6360046750343138","host":"s3://npm-registry-packages-npm-production"}},"1.4.2":{"name":"memorykit-mcp-server","version":"1.4.2","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","_id":"memorykit-mcp-server@1.4.2","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"homepage":"https://github.com/rapozoantonio/memorykit#readme","bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"bin":{"memorykit":"dist/index.js"},"dist":{"shasum":"2c850ca90e433152442a89a9f6ab66f1f6e788a3","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-1.4.2.tgz","fileCount":113,"integrity":"sha512-5H19PnPj7+XtiBESOUb4Qj0sLgSdzQr/M3R2aLZKX7ZflVZ6OQAnqV2wsctIWBl+kz5DlBg4XMkLLGZhyCyS6A==","signatures":[{"sig":"MEYCIQDXjtJGmP1unZFKFbs0ivvVgOEqFovMpMGm1J9ki+LxEAIhAOcaZHKmibrh/yLX0FEr4F2CdtG70dbl8Ve2mT8cl5g8","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":454172},"main":"./dist/index.js","type":"module","types":"./dist/index.d.ts","engines":{"node":">=18.0.0"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"gitHead":"7b52d512dfb32eb605782085960adb4b7bfb429e","mcpName":"io.github.rapozoantonio/memorykit","scripts":{"dev":"tsx src/index.ts","test":"vitest","build":"tsc","start":"node dist/index.js","prepare":"npm run build","test:ci":"vitest run","prepublishOnly":"npm run test:ci && npm run build"},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"overrides":{"sharp":"^0.33.0"},"repository":{"url":"git+https://github.com/rapozoantonio/memorykit.git","type":"git","directory":"mcp-server"},"_npmVersion":"11.6.2","description":"Cognitive memory for AI coding assistants — persistent memory across sessions","directories":{},"_nodeVersion":"24.12.0","dependencies":{"zod":"^3.22.0","glob":"^10.3.0","yaml":"^2.3.0","commander":"^11.0.0","@xenova/transformers":"^2.17.0","@modelcontextprotocol/sdk":"^1.0.0"},"_hasShrinkwrap":false,"devDependencies":{"tsx":"^4.7.0","vitest":"^1.0.0","typescript":"^5.3.0","@types/node":"^20.0.0"},"_npmOperationalInternal":{"tmp":"tmp/memorykit-mcp-server_1.4.2_1785422564622_0.6952654144679904","host":"s3://npm-registry-packages-npm-production"}},"2.0.0":{"name":"memorykit-mcp-server","version":"2.0.0","description":"Cognitive memory for AI coding assistants — persistent memory across sessions","mcpName":"io.github.rapozoantonio/memorykit","type":"module","main":"./dist/index.js","types":"./dist/index.d.ts","exports":{".":{"import":"./dist/index.js","types":"./dist/index.d.ts"}},"bin":{"memorykit":"dist/index.js"},"scripts":{"build":"tsc","dev":"tsx src/index.ts","start":"node dist/index.js","test":"vitest","test:ci":"vitest run","eval":"node evals/run.mjs","prepare":"npm run build","prepublishOnly":"npm run test:ci && npm run build"},"keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"author":{"name":"Antonio Rapozo"},"license":"MIT","repository":{"type":"git","url":"git+https://github.com/rapozoantonio/memorykit.git","directory":"mcp-server"},"bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"homepage":"https://github.com/rapozoantonio/memorykit#readme","engines":{"node":">=18.0.0"},"dependencies":{"@modelcontextprotocol/sdk":"^1.0.0","@xenova/transformers":"^2.17.0","commander":"^11.0.0","glob":"^10.3.0","yaml":"^2.3.0","zod":"^3.22.0"},"overrides":{"sharp":"^0.33.0"},"devDependencies":{"@types/node":"^20.0.0","tsx":"^4.7.0","typescript":"^5.3.0","vitest":"^1.0.0"},"gitHead":"56c20829244c1ba2d74b19be8cfac3924ba4883e","_id":"memorykit-mcp-server@2.0.0","_nodeVersion":"24.12.0","_npmVersion":"11.6.2","dist":{"integrity":"sha512-sLSUsywrjvbLMB2Ec44+HZOp8rWbf4mNrNeixFTVwSPb6sMTbfzks11WIYNIaLtXL8Ja62LIe4Vr6jnQ1PJ8QA==","shasum":"0533c4e24e85e7508ba0cfe126d692f6d0c1d730","tarball":"https://registry.npmjs.org/memorykit-mcp-server/-/memorykit-mcp-server-2.0.0.tgz","fileCount":117,"unpackedSize":473992,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCIBlExs/jR4Ga1aLMjnWKdL8p2M4KzFUoC4iX5pP/HLFKAiEA0prz3l5Y2MII+xrnjKezuRX/UofG15oN0VaNLIuDimM="}]},"_npmUser":{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"},"directories":{},"maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/memorykit-mcp-server_2.0.0_1788977874742_0.3701582602077511"},"_hasShrinkwrap":false}},"time":{"created":"2026-06-18T14:58:58.702Z","modified":"2026-09-09T18:17:55.044Z","1.0.0":"2026-06-18T14:58:59.080Z","1.0.1":"2026-06-19T00:16:34.326Z","1.1.0":"2026-07-14T16:05:17.438Z","1.1.1":"2026-07-20T15:54:14.438Z","1.2.0":"2026-07-21T12:24:04.581Z","1.3.0":"2026-07-22T09:34:48.527Z","1.3.1":"2026-07-22T13:48:43.154Z","1.4.0":"2026-07-30T13:23:50.496Z","1.4.1":"2026-07-30T14:33:19.005Z","1.4.2":"2026-07-30T14:42:44.756Z","2.0.0":"2026-09-09T18:17:54.885Z"},"bugs":{"url":"https://github.com/rapozoantonio/memorykit/issues"},"author":{"name":"Antonio Rapozo"},"license":"MIT","homepage":"https://github.com/rapozoantonio/memorykit#readme","keywords":["mcp","memory","ai","cursor","copilot","claude","llm","context","cognitive","model-context-protocol"],"repository":{"type":"git","url":"git+https://github.com/rapozoantonio/memorykit.git","directory":"mcp-server"},"description":"Cognitive memory for AI coding assistants — persistent memory across sessions","maintainers":[{"name":"rapozoantonio","email":"rapozoantonio@gmail.com"}],"readme":"# MemoryKit MCP Server\n\n**Cognitive memory for AI coding assistants** — gives Claude Desktop, Claude Code, GitHub Copilot, and Cursor persistent memory across conversations. No database, no Docker, no API keys required.\n\n---\n\n## The Memory That Pays for Itself\n\nMost memory tools store everything and make you pay full re-discovery cost on every retrieval. MemoryKit doesn't:\n\n- **Costs scale with the question, not your history** — a quick check-in costs ~200 tokens, a deep recall costs up to 1,500. Never the whole store.\n- **Junk never gets stored** — low-value and duplicate entries are rejected before they're written, so you're not paying to filter noise later.\n- **It shrinks itself** — stale entries auto-prune, important ones auto-promote. No manual cleanup, no slow bloat after months of use.\n- **You see the receipt** — every retrieval reports real numbers: tokens spent learning a fact vs. tokens spent recalling it.\n\n| You ask                    | Context budget |\n| -------------------------- | -------------- |\n| \"what was I doing?\"        | ~200 tokens    |\n| \"how do I deploy this?\"    | ~300 tokens    |\n| \"what's our DB choice?\"    | ~500 tokens    |\n| \"what happened last week?\" | ~1,500 tokens  |\n\n_Example: spend 1,200 tokens figuring something out once, recall it for ~70 tokens later — MemoryKit reports that 94% efficiency gain back to you, every time._\n\n## Accurate, Not Just Fast\n\n- **Understands meaning, not just keywords** — semantic search catches paraphrased questions a plain keyword match would miss.\n- **Recent + important wins over \"kinda similar\"** — stale notes don't outrank what actually matters right now.\n- **Catches contradictions** — conflicting info gets flagged, not silently duplicated.\n\n---\n\n## How It Works\n\nMemoryKit stores memories as Markdown files on your local filesystem using a brain-inspired 4-layer architecture:\n\n| Layer          | What it stores                             | Lifetime                              |\n| -------------- | ------------------------------------------ | ------------------------------------- |\n| **Working**    | Active session context, in-progress tasks  | Short-lived (decays after 7 days)     |\n| **Facts**      | Architecture decisions, tech stack choices | Permanent                             |\n| **Episodes**   | Bugs found, incidents, debugging sessions  | Medium-term (compacted after 30 days) |\n| **Procedures** | Coding rules, conventions, how-to guides   | Permanent                             |\n\nMemories are stored under `~/.memorykit/<project-name>/` — isolated per project via automatic git root detection. No configuration required for basic use.\n\n---\n\n## Quick Start\n\n### Claude Code: install as a plugin (recommended)\n\n```\n/plugin marketplace add rapozoantonio/memorykit\n/plugin install memorykit@memorykit\n```\n\nThis installs the MCP server, the SessionStart hook and the `/recall` and `/save` skills as one unit, and writes nothing into your repository. Skip to [Available Tools](#available-tools) — there is no `init` step.\n\nFor every other harness, or if you'd rather have the config files checked in, use `init` below.\n\n### 1. Initialize in your project\n\n```bash\ncd /your/project\nnpx -y memorykit-mcp-server@latest init\n```\n\nNo global install needed — `npx` always runs the latest published version, so you never have to remember to update. (If you'd rather pin a version or skip the network check on every launch, `npm install -g memorykit-mcp-server` still works; see [Keeping MemoryKit up to date](#keeping-memorykit-up-to-date).)\n\nThis creates:\n\n- `~/.memorykit/<project-name>/` — Memory storage directory\n- `.vscode/mcp.json` — GitHub Copilot MCP server config\n- `.mcp.json` — Claude Code MCP server config\n- `CLAUDE.md` — Claude Code instructions to use memory proactively\n- `.github/copilot-instructions.md` — GitHub Copilot instructions to use memory proactively\n- `.claude/settings.local.json` — SessionStart hook that recalls memory when a session starts, clears, or compacts\n- `.claude/skills/recall/`, `.claude/skills/save/` — `/recall` and `/save` slash commands\n\nThe instruction files tell AI models to automatically check memory before starting tasks and save learnings when completing work. This ensures memory is used consistently without manual prompting.\n\n### 2. Configure your AI assistant\n\n**GitHub Copilot in VS Code** — Already configured! `memorykit init` creates `.vscode/mcp.json` and `.github/copilot-instructions.md` automatically. The instructions tell Copilot to check memory before tasks and save learnings after.\n\n**Claude Code in VS Code** — Already configured! `memorykit init` creates `.mcp.json` and `CLAUDE.md` automatically. The instructions tell Claude to check memory before tasks and save learnings after.\n\n**Claude Desktop** — Edit the config file:\n\n| OS      | Path                                                              |\n| ------- | ----------------------------------------------------------------- |\n| Windows | `%APPDATA%\\Claude\\claude_desktop_config.json`                     |\n| macOS   | `~/Library/Application Support/Claude/claude_desktop_config.json` |\n| Linux   | `~/.config/Claude/claude_desktop_config.json`                     |\n\n```json\n{\n  \"mcpServers\": {\n    \"memorykit\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"memorykit-mcp-server@latest\"],\n      \"env\": {\n        \"MEMORYKIT_PROJECT\": \"/absolute/path/to/your/project\"\n      }\n    }\n  }\n}\n```\n\n**Cursor** — Add to Cursor MCP settings using the same format as Claude Desktop.\n\n### 3. Restart your AI assistant\n\nThe 7 MemoryKit tools will appear in the tool list:\n\n- `initialize_memory` — Create memory storage structure (run once per project)\n- `store_memory` — Save new memories\n- `retrieve_context` — Query relevant memories\n- `update_memory` — Modify existing entries\n- `forget_memory` — Delete entries\n- `list_memories` — Browse stored memories\n- `consolidate` — Manual cleanup/optimization (auto-runs every 5 minutes)\n\n---\n\n## Keeping MemoryKit up to date\n\nThe config examples above use `npx -y memorykit-mcp-server@latest` as the `command`/`args`. Every time your AI assistant launches the server, `npx` resolves `@latest` against the npm registry and runs that version — so you're always on the newest release without doing anything.\n\nIf you'd rather not pay the small `npx` resolution check on every launch (or want to pin a specific version for reproducibility), install globally instead and point `command` at the fixed binary:\n\n```bash\nnpm install -g memorykit-mcp-server\n```\n\n```json\n{\n  \"command\": \"memorykit\",\n  \"env\": { \"MEMORYKIT_PROJECT\": \"/absolute/path/to/your/project\" }\n}\n```\n\nWith a global install there's no auto-update — periodically run `npm install -g memorykit-mcp-server@latest` (or `npm outdated -g` to check) to pick up new releases.\n\n---\n\n## Available Tools\n\n### `store_memory`\n\nSave a new memory entry. Importance is scored automatically (0.1–0.95, never absolute 0 or 1) and the correct layer is selected based on content type.\n\n`why` is required. A memory that records what was decided without why it was decided retrieves poorly and cannot be acted on later — retrieval quality is bounded by what was written at store time.\n\n| Parameter             | Type     | Required | Description                                                             |\n| --------------------- | -------- | -------- | ----------------------------------------------------------------------- |\n| `content`             | string   | ✅       | The memory content                                                      |\n| `why`                 | string   | ✅       | The reason, root cause, or tradeoff behind it — not a restatement of `content` |\n| `tags`                | string[] | ❌       | Categorization tags (auto-detected if omitted)                          |\n| `layer`               | enum     | ❌       | `working`, `facts`, `episodes`, `procedures` (auto-detected if omitted) |\n| `scope`               | enum     | ❌       | `project` (default) or `global`                                         |\n| `file_hint`           | string   | ❌       | Target filename within layer (e.g. `\"technology\"`)                      |\n| `acquisition_context` | object   | ❌       | ROI tracking: `{ tokens_consumed: number, tool_calls: number }`         |\n\n**Example:**\n\n```json\n{\n  \"content\": \"We decided to use PostgreSQL as the primary database.\",\n  \"why\": \"ACID guarantees are required for billing, and the team already runs Postgres in production.\",\n  \"tags\": [\"database\", \"architecture\"],\n  \"acquisition_context\": { \"tokens_consumed\": 1200, \"tool_calls\": 3 }\n}\n```\n\n---\n\n### `retrieve_context`\n\nGet relevant memory context for a query. The Prefrontal Controller classifies your query and routes to the appropriate memory layers automatically.\n\n| Parameter    | Type     | Required | Description                                                      |\n| ------------ | -------- | -------- | ---------------------------------------------------------------- |\n| `query`      | string   | ✅       | Natural language question or topic                               |\n| `max_tokens` | number   | ❌       | Token budget override (default: query-type based — 200 to 2,000) |\n| `layers`     | string[] | ❌       | Restrict to specific layers                                      |\n| `scope`      | enum     | ❌       | `all` (default), `project`, or `global`                          |\n\n**Example:**\n\n```json\n{\n  \"query\": \"what database are we using and why?\",\n  \"scope\": \"all\"\n}\n```\n\n---\n\n### `update_memory`\n\nModify an existing memory entry by ID.\n\n| Parameter    | Type     | Required | Description                           |\n| ------------ | -------- | -------- | ------------------------------------- |\n| `entry_id`   | string   | ✅       | Entry ID to update                    |\n| `content`    | string   | ❌       | New content                           |\n| `tags`       | string[] | ❌       | Updated tags                          |\n| `importance` | number   | ❌       | Manual importance override (0.1–0.95) |\n\n---\n\n### `forget_memory`\n\nDelete a memory entry by ID.\n\n| Parameter  | Type   | Required | Description        |\n| ---------- | ------ | -------- | ------------------ |\n| `entry_id` | string | ✅       | Entry ID to delete |\n\n---\n\n### `consolidate`\n\nRun memory maintenance: prune stale working memory, promote high-importance entries to long-term layers, and compact old episode files.\n\n| Parameter | Type    | Required | Description                             |\n| --------- | ------- | -------- | --------------------------------------- |\n| `scope`   | enum    | ❌       | `project` (default), `global`, or `all` |\n| `dry_run` | boolean | ❌       | Preview changes without modifying files |\n\n---\n\n### `list_memories`\n\nBrowse the memory structure and see entry counts per layer.\n\n| Parameter | Type | Required | Description                             |\n| --------- | ---- | -------- | --------------------------------------- |\n| `scope`   | enum | ❌       | `all` (default), `project`, or `global` |\n| `layer`   | enum | ❌       | Filter to a specific layer              |\n\n---\n\n## Memory Quality Gates\n\nMemoryKit filters low-quality entries automatically before storing:\n\n1. **Importance Floor** — Content scoring below 0.15 is rejected as routine or trivial.\n2. **Duplicate Detection** — Near-duplicate entries are blocked; the existing entry ID is returned so you can update it instead.\n3. **Contradiction Warning** — If new content potentially conflicts with existing knowledge, storage proceeds but a warning is returned.\n\n---\n\n## CLI Commands\n\n```bash\n# Initialize memory for the current project\nmemorykit init\n\n# Initialize global memory shared across all projects\nmemorykit init --global\n\n# Show memory statistics (entry counts, file sizes, last consolidation)\nmemorykit status\n\n# Run memory consolidation (prune stale, promote important, compact old episodes)\nmemorykit consolidate\n\n# Preview consolidation without making changes\nmemorykit consolidate --dry-run\n```\n\n---\n\n## Configuration\n\nAfter `memorykit init`, a `memorykit.yaml` is created at `~/.memorykit/<project-name>/memorykit.yaml`:\n\n```yaml\nversion: \"0.1\"\n\nworking:\n  max_entries: 50\n  decay_threshold_days: 7\n  promotion_threshold: 0.70\n\nfacts:\n  max_entries_per_file: 100\n\nepisodes:\n  compaction_after_days: 30\n\nconsolidation:\n  auto: true\n  interval_minutes: 0\n\nglobal:\n  enabled: true\n  priority: \"project\"\n\ncontext:\n  max_tokens_estimate: 4000\n\nquality_gates:\n  importance_floor: 0.15\n  duplicate_jaccard_threshold: 0.6\n  duplicate_word_overlap: 3\n```\n\n---\n\n## Storage Layout\n\n```\n~/.memorykit/\n├── <project-name>/\n│   ├── memorykit.yaml\n│   ├── working/\n│   │   └── session.md\n│   ├── facts/\n│   │   ├── architecture.md\n│   │   ├── technology.md\n│   │   └── general.md\n│   ├── episodes/\n│   │   └── 2026-03-04.md\n│   └── procedures/\n│       └── general.md\n└── facts/                       # Global memory (shared across all projects)\n    └── ...\n```\n\n---\n\n## Architecture\n\n```\nAI Assistant (Claude / Copilot / Cursor)\n     │  MCP Protocol (stdio)\n     ↓\nMemoryKit MCP Server (Node.js)\n     ├── Prefrontal Controller   Query classification & file routing\n     ├── Amygdala Engine         Importance scoring (9-signal, 0.1–0.95)\n     ├── Quality Gates           Importance floor, duplicate detection, contradiction warning\n     ├── Normalizer              Prose-to-MML normalization pipeline\n     └── File Storage            Local Markdown files (~/.memorykit/)\n```\n\n**Query Classification** (Prefrontal): Routes `retrieve_context` queries to the right files:\n\n- _Continuation_ → `working/session.md`\n- _Fact retrieval_ → `facts/*.md`\n- _Deep recall_ → `episodes/*.md`\n- _Procedural_ → `procedures/*.md`\n- _Complex_ → all layers\n\n**Importance Scoring** (Amygdala): 9 signals scored 0.1–0.95 — decision language, explicit importance markers, code blocks, technical depth, novelty, sentiment, conversation context, question patterns, and MML structure.\n\n---\n\n## Environment Variables\n\n| Variable            | Description                   | Default                     |\n| ------------------- | ----------------------------- | --------------------------- |\n| `MEMORYKIT_PROJECT` | Absolute path to project root | Auto-detected from git root |\n\n---\n\n## Requirements\n\n- Node.js 18+\n- No database required\n- No Docker required\n- No API keys required\n\n---\n\n## Contributing\n\nSee [CONTRIBUTING.md](../CONTRIBUTING.md).\n\n## Changelog\n\nSee [CHANGELOG.md](CHANGELOG.md).\n\n## License\n\nMIT — see [LICENSE](LICENSE) for details.\n\n## Support\n\n- Issues: [GitHub Issues](https://github.com/antonio-rapozo/memorykit/issues)\n- Discussions: [GitHub Discussions](https://github.com/antonio-rapozo/memorykit/discussions)\n","readmeFilename":"README.md"}