{"_id":"@dzhechkov/skills-bto","_rev":"17-3657d593d9f634e2f64df8d5e2021e5f","name":"@dzhechkov/skills-bto","dist-tags":{"latest":"1.4.7"},"versions":{"1.0.0":{"name":"@dzhechkov/skills-bto","version":"1.0.0","keywords":["claude","claude-code","ai","bto","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.0.0","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/dzhechko/product-keysarium-2026#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"bd46354803e6f40e377a5bc2b5dda4220ba756da","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.0.0.tgz","fileCount":25,"integrity":"sha512-/gjC4+sA9WK/uwi0pCQBYGVmDzq0NLWguNww+7AO8yopUynp3t3lrSJJlIrT48rZscZFhZnfbIu9bjW0Qt4Hmw==","signatures":[{"sig":"MEYCIQDA4p+f3z6VFa8cLpdTLm1JvBs2IbHS57m4s5NZBIbAOAIhAKT6NdivnHCwt4UNOVd8clny7UpJTSHPBqFHvqYm5jkc","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":180741},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"38bfafc083f43521be562a3eeaea609f671988ce","scripts":{"test":"node bin/cli.js doctor","prepublishOnly":"node scripts/sync-templates.js"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/dzhechko/product-keysarium-2026.git","type":"git"},"_npmVersion":"11.6.2","description":"Build-Test-Optimize skill pack for Claude Code — structured BTO pipeline with quality gates, testing strategies, and optimization workflows","directories":{},"_nodeVersion":"24.11.1","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.0.0_1772391763043_0.6248071429983042","host":"s3://npm-registry-packages-npm-production"}},"1.0.1":{"name":"@dzhechkov/skills-bto","version":"1.0.1","keywords":["claude","claude-code","ai","bto","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.0.1","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/dzhechko/product-keysarium-2026#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"802c742a83714a8e107f47d8d42788f6236e024d","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.0.1.tgz","fileCount":26,"integrity":"sha512-jZ/10staqXbchnt4Xad+t6SMmXqL+UNweFIH2r3gAtqJuLBYW6aZwUDVF4MDYSiEOEhnlvy4luoIGyRCld1Pzw==","signatures":[{"sig":"MEUCIQCNqrOOTTj+G7s5OtcIT7p1ax6YXkyA+MYv3QoB2iy6aQIgEFnUxGHXH6cq0QospMEjh76HJXY1nlMOFnqOcJK2LyQ=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":186008},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"38bfafc083f43521be562a3eeaea609f671988ce","scripts":{"test":"node bin/cli.js doctor","prepublishOnly":"node scripts/sync-templates.js"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/dzhechko/product-keysarium-2026.git","type":"git"},"_npmVersion":"11.6.2","description":"Build-Test-Optimize skill pack for Claude Code — structured BTO pipeline with quality gates, testing strategies, and optimization workflows","directories":{},"_nodeVersion":"24.11.1","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.0.1_1772391912818_0.6996825556378483","host":"s3://npm-registry-packages-npm-production"}},"1.1.0":{"name":"@dzhechkov/skills-bto","version":"1.1.0","keywords":["claude","claude-code","ai","bto","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.1.0","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/dzhechko/product-keysarium-2026#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"fe9969c09a6eb5745b52840b0cbf4c70800c7ffb","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.1.0.tgz","fileCount":27,"integrity":"sha512-FFjzJaPN075JqkMHjGtegYDcOhi5M4zut/i+kyMBWv2mcFVjfke3i5qbIWCTJMKhKOCFYb4qXaQJxLthjz7flg==","signatures":[{"sig":"MEUCIAte/S9HpZHmu38U7AC3HCrgG6Sa6z+F/1UQs0vbaRzOAiEA/47/pobqA1QQQoRaVpJkK/9cQFTTrDlsd5hV7BaOOww=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":187743},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"76cd2c962aa75792c03b32c80e539b879831e000","scripts":{"test":"node bin/cli.js doctor","prepublishOnly":"node scripts/sync-templates.js"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/dzhechko/product-keysarium-2026.git","type":"git"},"_npmVersion":"11.6.2","description":"Build-Test-Optimize skill pack for Claude Code — structured BTO pipeline with quality gates, testing strategies, and optimization workflows","directories":{},"_nodeVersion":"24.11.1","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.1.0_1772395888637_0.13354639017116865","host":"s3://npm-registry-packages-npm-production"}},"1.1.1":{"name":"@dzhechkov/skills-bto","version":"1.1.1","keywords":["claude","claude-code","ai","bto","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.1.1","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/dzhechko/product-keysarium-2026#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"cf72e6f69bac4364e0358fd8f99f6a5a5d0d4714","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.1.1.tgz","fileCount":27,"integrity":"sha512-kZedYtyeshJ9O/tAxAch6CfucruSCgm7IU86GYlzbH5/gIsgRzQ8OlGmmr0Y2fYCW04IFa04vz0OIQ/aQJVr0w==","signatures":[{"sig":"MEUCIQDbKhw1VWAcsu8Xy4Euj6yh1MFfVm26ooWVJkwCCeJYRgIgJrtTR7nAjSlulSbbMm+N0X7TQfRJNeabbGdlasi1BvU=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":187862},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"990895fccb23c0bbef6ef0714436bf9e6706b256","scripts":{"test":"node bin/cli.js doctor","prepublishOnly":"node scripts/sync-templates.js"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/dzhechko/product-keysarium-2026.git","type":"git"},"_npmVersion":"11.6.2","description":"Build-Test-Optimize skill pack for Claude Code — structured BTO pipeline with quality gates, testing strategies, and optimization workflows","directories":{},"_nodeVersion":"24.11.1","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.1.1_1772396023537_0.1674439722923864","host":"s3://npm-registry-packages-npm-production"}},"1.2.0":{"name":"@dzhechkov/skills-bto","version":"1.2.0","keywords":["claude","claude-code","ai","bto","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.2.0","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/dzhechko/product-keysarium-2026#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"bb3a081d45644b264dac34bda0dcd8db20903e10","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.2.0.tgz","fileCount":31,"integrity":"sha512-zD1mfYbREdPPITisj7DQLEwSEMngTqCJdPoCckG5eAGpk/gbcmS3R2uSmH3vQmjTqs0G56qbiwFL+R3Qy+e3YQ==","signatures":[{"sig":"MEUCIQC2ZjbKgproGk+cjXF7SrVuum5SRiIu3L5vCRZ3Aa3OuAIgFRmwG6Yn4Oqu1EPTmXglmMXWG7R9CbuseytJb7jMavE=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":223171},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"b81f9ef3221d3384c53ef0be333970afe59b9fa2","scripts":{"test":"node bin/cli.js doctor","prepublishOnly":"node scripts/sync-templates.js"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/dzhechko/product-keysarium-2026.git","type":"git"},"_npmVersion":"11.6.2","description":"Build-Test-Optimize skill pack for Claude Code — structured BTO pipeline with quality gates, witness chain verification, judge attestation, and optimization workflows","directories":{},"_nodeVersion":"24.11.1","_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.2.0_1772431503749_0.6874386927268854","host":"s3://npm-registry-packages-npm-production"}},"1.3.0":{"name":"@dzhechkov/skills-bto","version":"1.3.0","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.3.0","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/dzhechko/product-keysarium-2026#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"e2cf720889e3311155a973d610103b26b040b969","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.3.0.tgz","fileCount":35,"integrity":"sha512-jY1sOcC/egZdKkWIK2CKOk37T1qHOSwPcn+GV9DVWiKhKatQ6OxEq3DnCRhTLlcalV0I5pLbRpPIxizEoUhX5w==","signatures":[{"sig":"MEYCIQDEpxC73dVPXwNfZ2hiyMdUfYqXyngQJgR1qD0o29YENAIhAOjuPnfTHe0C5CJgjW+S5X0VL2J23MklQpAsgJYJMYRy","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":308451},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"76c09b27dbd13f0eb1bf4bfb4ac3473d10a145df","scripts":{"test":"node bin/cli.js doctor","prepublishOnly":"node scripts/sync-templates.js"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/dzhechko/product-keysarium-2026.git","type":"git"},"_npmVersion":"11.6.2","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"24.11.1","_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.3.0_1772547597807_0.24594909615775973","host":"s3://npm-registry-packages-npm-production"}},"1.3.1":{"name":"@dzhechkov/skills-bto","version":"1.3.1","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.3.1","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"978cdf0a6933cc210c2dcb88e13bb3c0d076a910","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.3.1.tgz","fileCount":35,"integrity":"sha512-lAOx63MzHI06YcIN1kV1axODP+PrcSzxfnTRbBzXfAZZT/vzWVBduSBh+GU9/D+nxfmEPzud1oAn+0Nl9Hoe6g==","signatures":[{"sig":"MEQCIFB4gDalNUUoGwzie7VFWP3SGw8JMCwe/Nhu+QSaER31AiAnVIMSK41G8uh8+NiA0Izdk3eoN4QTm6jVkZbkZc35lw==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":310935},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"e84b19b97cdd7bc06a3011c50562d3f24b7c0403","_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.3.1_1781207828563_0.6559638981379621","host":"s3://npm-registry-packages-npm-production"}},"1.3.2":{"name":"@dzhechkov/skills-bto","version":"1.3.2","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.3.2","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"6d1042bbfc05543cf6181b86acd56d4ecd531f20","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.3.2.tgz","fileCount":36,"integrity":"sha512-N9fafceXxoF7FcapllQsCby+XKAhV/MqnbiwyUOJL1D04AOARDwuqr/rpHW8//QyIpWURn148aAaFtGCKqNCsw==","signatures":[{"sig":"MEYCIQChfk1Zi26bUwZsbGls007WUhLIe5UkWCxS7nLVVWrBLgIhAOSJdM+/kNWZ1TWbMuKkdkmM59tZHvVJBPW2DG9iAaXR","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":312015},"main":"src/cli.js","engines":{"node":">=16.0.0"},"gitHead":"b3b18605ef456b46f1ebc81b6c69b8e46e620aed","_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.3.2_1781211762749_0.25633177424846676","host":"s3://npm-registry-packages-npm-production"}},"1.3.3":{"name":"@dzhechkov/skills-bto","version":"1.3.3","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.3.3","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"0292edaa9fca19a5eb7383fb69aca91b342fb253","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.3.3.tgz","fileCount":36,"integrity":"sha512-Zxk1wYTx0uqLVzcxJyAtvOPLNILVN+9WMawE2t8YodrXOuAwSa3H1tIr3IQGOXVKFovWxSAGZBm5I3MPxKW1Jg==","signatures":[{"sig":"MEUCIQC+WdDsmJuzesmCOXTjHA4dGMWz1DZN6ucgrPkw7UA7XgIgVWEX3GRtkA7HAfuPDlb5smHwnFfG0mjzicGvi6y9mbg=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":314580},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.3.3.tgz","engines":{"node":">=16.0.0"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/964919477514133f88447932fee9041f/dzhechkov-skills-bto-1.3.3.tgz","_integrity":"sha512-Zxk1wYTx0uqLVzcxJyAtvOPLNILVN+9WMawE2t8YodrXOuAwSa3H1tIr3IQGOXVKFovWxSAGZBm5I3MPxKW1Jg==","repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.3.3_1781638089639_0.08277708046170651","host":"s3://npm-registry-packages-npm-production"}},"1.3.4":{"name":"@dzhechkov/skills-bto","version":"1.3.4","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.3.4","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"442bab30267205d4c43fe4aca6b164fd5640bd61","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.3.4.tgz","fileCount":36,"integrity":"sha512-9CbfGI8xX9bd1DhC6QNgx9zEicbJE3J13L59EIKYkSVCev19IF2T7K3q9DZfB1+BH8+XfALRKygLZUsHXtnorg==","signatures":[{"sig":"MEQCIHAPVzmgTgOk3ljn9zQpzuoLeoiiMcOSDRmbeJ7/4eDAAiBEVipjJBS8kujwMoHdI3oE8L/PHUokMDhZtKCWXklsbw==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":318406},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.3.4.tgz","engines":{"node":">=16.0.0"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/02256b0e6a28805d722547bdaddaa4f1/dzhechkov-skills-bto-1.3.4.tgz","_integrity":"sha512-9CbfGI8xX9bd1DhC6QNgx9zEicbJE3J13L59EIKYkSVCev19IF2T7K3q9DZfB1+BH8+XfALRKygLZUsHXtnorg==","repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.3.4_1784118658925_0.47665802010740865","host":"s3://npm-registry-packages-npm-production"}},"1.4.1":{"name":"@dzhechkov/skills-bto","version":"1.4.1","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.4.1","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"0aa5529d5404cd25a9ed9a0761767d91e8974ca4","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.4.1.tgz","fileCount":39,"integrity":"sha512-uj1T80QhzYFntVnFDVKWTIsJ9BdnaECPxkfnOXedcLZBakGzB9WCkyw+XOTFbL+2SDnCJmldeNagB4ZoFoV75w==","signatures":[{"sig":"MEUCIEzlw03IhV7uKlbbxw3efduTja1LOXop5zhX1m/wFt7eAiEAmu8hyQOuQfLFkZinhy5bjSsL4kqKaIbv6fgYvyjsAqo=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":373748},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.4.1.tgz","engines":{"node":">=16.0.0"},"scripts":{"test":"node --test test/*.test.mjs"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/8e2286825a615418d7cf2f0c20e22fa9/dzhechkov-skills-bto-1.4.1.tgz","_integrity":"sha512-uj1T80QhzYFntVnFDVKWTIsJ9BdnaECPxkfnOXedcLZBakGzB9WCkyw+XOTFbL+2SDnCJmldeNagB4ZoFoV75w==","repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.4.1_1787632925811_0.44179608327960174","host":"s3://npm-registry-packages-npm-production"}},"1.4.2":{"name":"@dzhechkov/skills-bto","version":"1.4.2","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.4.2","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"101e3cf3a6456ab48a46f057cbf3aa969ddd4ff4","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.4.2.tgz","fileCount":39,"integrity":"sha512-Hy1larBP+JFw+pDaFmWcI4EVSHiP3aAthGGALZktwggUAdRcTqJOs2FKgmE9QQFxy/57yqC+IBxj7wfh62D+Gw==","signatures":[{"sig":"MEYCIQD8FbSc14DBx2BSSENSarU8Y+Q6a5x23EfuBPOdPdTSZAIhAKUy1RvQfU7lJ5qc7zzpS1FF1hJq/RYcvA0dyBhdAPfJ","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":373885},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.4.2.tgz","engines":{"node":">=16.0.0"},"scripts":{"test":"node --test test/*.test.mjs"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/5ec971b34e7a4805b1327fd83d4e45b9/dzhechkov-skills-bto-1.4.2.tgz","_integrity":"sha512-Hy1larBP+JFw+pDaFmWcI4EVSHiP3aAthGGALZktwggUAdRcTqJOs2FKgmE9QQFxy/57yqC+IBxj7wfh62D+Gw==","repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.4.2_1787944500212_0.37504752033010313","host":"s3://npm-registry-packages-npm-production"}},"1.4.3":{"name":"@dzhechkov/skills-bto","version":"1.4.3","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.4.3","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"b04ea5b2ca6a7dd539819ef1d3e3576ff11688ad","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.4.3.tgz","fileCount":39,"integrity":"sha512-TG/S69G7aScsvUiA5JE5sBA+KXcAHShPJvynbH2Q9B4g0dOJK7QxBZ3HEf/qg1lQOFbov+E6pTu+3ses2FgDDg==","signatures":[{"sig":"MEUCIQChAjmW3kJ70aw2Let8/OKfQM+RKnIX6rCepjU4liXFgQIgK60BvSOC7Q3yENG+6cY6OWICKqhllE65DzDIS7bl+5w=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":373885},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.4.3.tgz","engines":{"node":">=16.0.0"},"scripts":{"test":"node --test test/*.test.mjs"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/a892b64c459a97f0daeedff96d800544/dzhechkov-skills-bto-1.4.3.tgz","_integrity":"sha512-TG/S69G7aScsvUiA5JE5sBA+KXcAHShPJvynbH2Q9B4g0dOJK7QxBZ3HEf/qg1lQOFbov+E6pTu+3ses2FgDDg==","repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.4.3_1787944881665_0.9458669592567517","host":"s3://npm-registry-packages-npm-production"}},"1.4.4":{"name":"@dzhechkov/skills-bto","version":"1.4.4","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.4.4","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness-hub/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/dzhechko/product-keysarium-2026/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"e7a86140aa81be232f64cbe8c86b880f4990ee99","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.4.4.tgz","fileCount":39,"integrity":"sha512-nbZidemPKaDR6qPNFCi89O/idZC65j/aGWipu4F/YdcctPjBlQZ8kLERWvUUQQLTjakGeM4MXCWEhxE+HUyRwA==","signatures":[{"sig":"MEUCIHj5/JXv3Zo99pEqn7F7m0hmDKAbGX7GcGPpcmaHwoiOAiEA1tHX8BaDOC0SCrO5BzjqEdcXl4KqEfIrZepQDKzuVYE=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":373885},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.4.4.tgz","engines":{"node":">=16.0.0"},"scripts":{"test":"node --test test/*.test.mjs"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/f91e83f90bffba1586e4496e3b8adada/dzhechkov-skills-bto-1.4.4.tgz","_integrity":"sha512-nbZidemPKaDR6qPNFCi89O/idZC65j/aGWipu4F/YdcctPjBlQZ8kLERWvUUQQLTjakGeM4MXCWEhxE+HUyRwA==","repository":{"url":"git+https://github.com/djd1m/dz-harness-hub.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.4.4_1788003791431_0.9717485486518087","host":"s3://npm-registry-packages-npm-production"}},"1.4.5":{"name":"@dzhechkov/skills-bto","version":"1.4.5","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.4.5","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/djd1m/dz-harness/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"71ce4e0d732b0cfd2bece94c4393fa72963af3ec","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.4.5.tgz","fileCount":39,"integrity":"sha512-Rko1RyfpumTwH9kzfxQwlL9rfxYf2+uyCSwWcZo3lzybe78KtLDqfYXhWYyOb8y8LLTz+JSSXEfrwyWjAWJ26A==","signatures":[{"sig":"MEQCICRJCMmCOumWJRZtJSKR2pItPMw/ZSm5CtlpGm4QsOJyAiBIK69j38pDjRNpuArTXHvMjGtoD+WQjRSyoGtCpQaFgw==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":373866},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.4.5.tgz","engines":{"node":">=16.0.0"},"scripts":{"test":"node --test test/*.test.mjs"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/e389cac886ab7ad9e9306a54fd281822/dzhechkov-skills-bto-1.4.5.tgz","_integrity":"sha512-Rko1RyfpumTwH9kzfxQwlL9rfxYf2+uyCSwWcZo3lzybe78KtLDqfYXhWYyOb8y8LLTz+JSSXEfrwyWjAWJ26A==","repository":{"url":"git+https://github.com/djd1m/dz-harness.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.4.5_1788025738682_0.32149194662353553","host":"s3://npm-registry-packages-npm-production"}},"1.4.6":{"name":"@dzhechkov/skills-bto","version":"1.4.6","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"author":{"name":"dzhechko"},"license":"MIT","_id":"@dzhechkov/skills-bto@1.4.6","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"homepage":"https://github.com/djd1m/dz-harness/tree/main/packages/@dzhechkov/skills-bto#readme","bugs":{"url":"https://github.com/djd1m/dz-harness/issues"},"bin":{"skills-bto":"bin/cli.js"},"dist":{"shasum":"6d180b82d557a2343e193ea8a022d2a51e1c73a7","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.4.6.tgz","fileCount":39,"integrity":"sha512-pITYCguHgVoCweZMkZmDWupK+KQvM8sZpSIu+GVQf/2fuxJkFTA4lKutcoGIS5VQQTcgJFtkaxJGOkX0TNEzpw==","signatures":[{"sig":"MEUCIQDus6s3jWXBxXXc7RAFDZKVbbN3LxCLRXTKFCMDC60AQgIgDzyNHDG3Qh2phKbXa6nrvff3bqIXi1CuxAKNt4sBbKU=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"sig":"MEYCIQCYVQrCAHInOsaw5c3rKBEP1dV+8Em3hIscyt6ky0arlwIhAMrH4INFHuv2vktA9IDM9YE9sJFZNiHwPZB66SphhZiT","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":373866},"main":"src/cli.js","_from":"file:dzhechkov-skills-bto-1.4.6.tgz","engines":{"node":">=16.0.0"},"scripts":{"test":"node --test test/*.test.mjs"},"_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"_resolved":"/tmp/0a4d86990a183f5811c448dd20067d39/dzhechkov-skills-bto-1.4.6.tgz","_integrity":"sha512-pITYCguHgVoCweZMkZmDWupK+KQvM8sZpSIu+GVQf/2fuxJkFTA4lKutcoGIS5VQQTcgJFtkaxJGOkX0TNEzpw==","repository":{"url":"git+https://github.com/djd1m/dz-harness.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"tmp":"tmp/skills-bto_1.4.6_1789293678995_0.5911119940803768","host":"s3://npm-registry-packages-npm-production"}},"1.4.7":{"_id":"@dzhechkov/skills-bto@1.4.7","bin":{"skills-bto":"bin/cli.js"},"bugs":{"url":"https://github.com/djd1m/dz-harness/issues"},"dist":{"shasum":"37f7e42ac06473b299c27a1f5f88c8cdf2689a95","tarball":"https://registry.npmjs.org/@dzhechkov/skills-bto/-/skills-bto-1.4.7.tgz","fileCount":39,"integrity":"sha512-eh2z21imEj0vx7ss4Sm1c6ebmgKlXj1qVHK4IZIGM0f43hgmYLFlJzoPf9/GO5mMeEdPs3sSmyG6ikprhRnHAg==","signatures":[{"sig":"MEYCIQC02jfSYfYnDSUpNkHVkMsCKvvMy5i0d4RcNa72mIuptQIhAPfeE/cpcwx0DZ7rrWynaJWVKJA/AWGi5/BQ5JkSDgWG","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEQCIH6TbckUx1f5mQqXLU71704R3Pvamm7HqiVxgz6NblGaAiB1my8UmzYG3wZN8wO2h8t6kFLvbAErDSa6+6BQnTXQLw=="}],"unpackedSize":373807},"main":"src/cli.js","name":"@dzhechkov/skills-bto","_from":"file:/var/tmp/dz-publish-packed-MUHRnz/dzhechkov-skills-bto-1.4.7.tgz","author":{"name":"dzhechko"},"engines":{"node":">=16.0.0"},"license":"MIT","scripts":{"test":"node --test test/*.test.mjs"},"version":"1.4.7","_npmUser":{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"},"homepage":"https://github.com/djd1m/dz-harness/tree/main/packages/@dzhechkov/skills-bto#readme","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"_resolved":"/var/tmp/dz-publish-packed-MUHRnz/dzhechkov-skills-bto-1.4.7.tgz","_integrity":"sha512-eh2z21imEj0vx7ss4Sm1c6ebmgKlXj1qVHK4IZIGM0f43hgmYLFlJzoPf9/GO5mMeEdPs3sSmyG6ikprhRnHAg==","repository":{"url":"git+https://github.com/djd1m/dz-harness.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"_npmVersion":"10.9.4","description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","directories":{},"maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"_nodeVersion":"22.22.0","publishConfig":{"access":"public"},"_hasShrinkwrap":false,"peerDependencies":{"@dzhechkov/keysarium-core":"^1.0.0"},"peerDependenciesMeta":{"@dzhechkov/keysarium-core":{"optional":true}},"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/skills-bto_1.4.7_1789936426203_0.983059055501867"}}},"time":{"created":"2026-03-01T19:02:42.937Z","modified":"2026-09-20T20:33:46.495Z","1.0.0":"2026-03-01T19:02:43.199Z","1.0.1":"2026-03-01T19:05:12.962Z","1.1.0":"2026-03-01T20:11:28.797Z","1.1.1":"2026-03-01T20:13:43.698Z","1.2.0":"2026-03-02T06:05:03.936Z","1.3.0":"2026-03-03T14:19:57.949Z","1.3.1":"2026-06-11T19:57:08.708Z","1.3.2":"2026-06-11T21:02:42.912Z","1.3.3":"2026-06-16T19:28:09.785Z","1.3.4":"2026-07-15T12:30:59.092Z","1.4.1":"2026-08-25T04:42:05.952Z","1.4.2":"2026-08-28T19:15:00.378Z","1.4.3":"2026-08-28T19:21:21.839Z","1.4.4":"2026-08-29T11:43:11.583Z","1.4.5":"2026-08-29T17:48:58.821Z","1.4.6":"2026-09-13T10:01:19.094Z","1.4.7":"2026-09-20T20:33:46.329Z"},"bugs":{"url":"https://github.com/djd1m/dz-harness/issues"},"author":{"name":"dzhechko"},"license":"MIT","homepage":"https://github.com/djd1m/dz-harness/tree/main/packages/@dzhechkov/skills-bto#readme","keywords":["claude","claude-code","ai","bto","benchmark","build-test-optimize","skills","quality-gates","testing","optimization","skill-pack"],"repository":{"url":"git+https://github.com/djd1m/dz-harness.git","type":"git","directory":"packages/@dzhechkov/skills-bto"},"description":"Build-Benchmark-Test-Optimize skill pack for Claude Code — deterministic benchmarking, quality gates, witness chain, judge attestation, and optimization","maintainers":[{"name":"dzhechkov","email":"jechkov.dmitriy@gmail.com"}],"readme":"# @dzhechkov/skills-bto\n\n**Build-Benchmark-Test-Optimize skill pack for Claude Code**\n\nMulti-agent evaluation and iterative optimization pipeline for Claude Code skills, commands, and prompts. Includes deterministic benchmarking with golden sample comparison, test suites, consistency probes, and performance metrics. Part of the [Keysarium](https://www.npmjs.com/package/@dzhechkov/keysarium) ecosystem.\n\n---\n\n## Quick Start\n\n```bash\n# One-command install via npx\nnpx @dzhechkov/skills-bto\n\n# Or install globally\nnpm install -g @dzhechkov/skills-bto\nskills-bto init\n\n# Install into a project that already has @dzhechkov/keysarium\nnpx @dzhechkov/skills-bto init\n```\n\nAfter installation, open Claude Code in your project directory and start using BTO commands.\n\n---\n\n## What You Get\n\n| Component | Count | Description |\n|-----------|-------|-------------|\n| **Skill** | 1 | `bto` — core Build-Benchmark-Test-Optimize skill with 4 modules |\n| **Commands** | 5 | `/bto`, `/bto-build`, `/bto-benchmark`, `/bto-test`, `/bto-optimize` |\n| **Rules** | 1 | `bto-quality-gates` — quality gate enforcement (incl. benchmark gates) |\n| **Shards** | 1 | `bto-evaluation` — context shard for BTO evaluation pipeline |\n| **Agent Templates** | 2 | `bto-judge-panel`, `bto-optimizer-worker` |\n| **References** | 5 | Eval patterns, judge rubrics, optimization methods, quality checklist, golden samples |\n| **Examples** | 2 | Sample evaluation report, sample benchmark report |\n\nEverything is installed into your project's `.claude/` directory and works natively with Claude Code.\n\n---\n\n## Commands\n\n```bash\nnpx @dzhechkov/skills-bto                    # Full install (interactive, same as init)\nnpx @dzhechkov/skills-bto init               # Install all components\nnpx @dzhechkov/skills-bto init --force       # Overwrite existing files\nnpx @dzhechkov/skills-bto init --dry-run     # Preview without making changes\nnpx @dzhechkov/skills-bto update             # Update to latest version\nnpx @dzhechkov/skills-bto remove             # Clean uninstall\nnpx @dzhechkov/skills-bto list               # Show installed components\nnpx @dzhechkov/skills-bto doctor             # Health check\n```\n\n---\n\n## BTO Pipeline\n\n```\nBUILD ──→ BENCHMARK ──→ TEST ──→ OPTIMIZE\n  │           │           │         │\n  │           │           │         └── Evolutionary mutation + re-evaluation (3 rounds)\n  │           │           └── Multi-layer evaluation: Layer 0 → Layer 1 → Layer 2\n  │           └── Deterministic benchmarking: golden samples, test suite, consistency, metrics\n  └── Generate skill/command from description\n```\n\n### Usage in Claude Code\n\n```bash\n# Full BTO cycle: build → benchmark → test → optimize\n/bto Create a skill for code review automation\n\n# Build only — generate a new skill or command\n/bto-build Create a skill that analyzes git commit patterns\n\n# Benchmark only — deterministic benchmarking against golden samples\n/bto-benchmark .claude/skills/my-skill/SKILL.md\n\n# Test only — evaluate an existing artifact\n/bto-test .claude/skills/my-skill/SKILL.md\n\n# Optimize only — iteratively improve an artifact\n/bto-optimize .claude/skills/my-skill/SKILL.md\n```\n\n---\n\n## Evaluation Architecture\n\n### Benchmark Layers (deterministic, pre-TEST)\n\n| Layer | Cost | Purpose |\n|-------|------|---------|\n| **B-1** | Zero (deterministic) | **Environment preconditions** — probes that each downstream layer *can* run, and records the answer |\n| **B0** | Zero (deterministic) | Golden sample comparison — section coverage, ordering, proportions |\n| **B1** | Zero (deterministic) | Deterministic test suite — 6 tests for skills, 5 per other artifact type, PASS/FAIL |\n| **B2** | Minimal (3× haiku) | Consistency probe — 3 parallel agents, agreement measurement |\n| **B3** | Zero (deterministic) | Performance metrics — token efficiency, bloat detection, redundancy |\n\n**Scoring:** `BENCHMARK = B0×0.30 + B1×0.35 + B2×0.15 + B3×0.20`\n(pass rate denominator = checks that actually **executed**, never the declared total)\n\n**Gate**, first matching row wins:\n\n| Condition | Verdict |\n|-----------|---------|\n| B-1 returned ABORT | **ABORT** — no score is emitted at all |\n| Any layer INCONCLUSIVE | **INCONCLUSIVE** — not a pass and not a failure; TEST is not entered |\n| Score < 0.50 | BLOCK |\n| Score 0.50–0.70 | WARN |\n| Score > 0.70 | PASS → proceed to TEST |\n\nLayer B-1 prints `Preconditions: ALL_GREEN | DEGRADED(list) | ABORT` in the header of **every**\nreport, clean runs included — an absent line is itself a finding. A layer the operator deliberately\nskipped via `--level` is renormalized away; a layer that was *requested and could not run* is\nINCONCLUSIVE and is never renormalized, because redistributing its weight can only raise the score.\n\n### TEST Layer Model\n\n| Layer | Agents | Model | Purpose |\n|-------|--------|-------|---------|\n| **Layer 0** | 0 | — | Deterministic pre-checks (structure, completeness, encoding) |\n| **Layer 1** | 1 | haiku | Fast semantic evaluation across 5 dimensions |\n| **Layer 2** | 3 | sonnet | Full judge panel: Domain Expert + Critic + Completeness Auditor |\n| **Meta** | 1 | opus | Disagreement resolution (triggered when score delta > 3) |\n\n### Judge Panel\n\n- **3 independent judges** evaluate each artifact in isolation\n- Judges never see each other's scores before submitting\n- Standard weights: Domain Expert (0.4) / Critic (0.3) / Completeness Auditor (0.3)\n- If `max_score - min_score > 3` → meta-judge escalation\n- Every evaluation report always emits three provenance lines — `Authored by:`, `Judged by:` and\n  `Cross-family: YES|NO` — so a report read six months later says whether it was cross-checked:\n\n  ```\n  Authored by:  claude-opus (anthropic)\n  Judged by:    Expert=claude-sonnet | Critic=gpt-5.5 | Auditor=claude-sonnet\n  Cross-family: YES (Critic=openai — different family than the author)\n  ```\n\n- Running the Critic seat on a **different model family** than the author is *recommended*; when no\n  second family is reachable the report prints a loud `SAME-FAMILY PANEL` banner saying CORRECTNESS\n  and ROBUSTNESS are self-assessment. The family axis is **advisory** — it changes no score and\n  blocks nothing. The separate *model-identity* rule still BLOCKS: the same model must not both\n  generate and judge an artifact.\n\n### Agent Authoring Rule\n\nEvery BTO agent that writes a document writes it **incrementally**: a small `Write` carrying the\ntitle and section skeleton first, then one `Edit` per section. An agent that emits no tool event for\nthe harness watchdog window (180 s inside a workflow runner, 600 s for a directly-spawned subagent)\nis killed, and a killed agent loses *all* unwritten output — there is no partial save. Genuinely\nshort emissions (the B2 probe answer, a one-line variant score) are exempt, and each exemption is\nwritten down at its site rather than inferred from its absence.\n\n### Quality Gates\n\n- **BENCHMARK** must pass (score ≥ 0.50) before TEST begins\n- A check or layer that could not execute is reported **INCONCLUSIVE** — never counted as a pass,\n  never silently dropped (`SKIPPED is not PASSED`, universal check U-13)\n- BENCHMARK score < 0.50 → BLOCK (artifact needs rework)\n- Layer 0 must pass before Layer 1\n- Layer 1 must pass before Layer 2\n- Optimization accepted only if `new_score - prev_score > 0.5`\n- 3 consecutive iterations with delta ≤ 0.5 → convergence declared\n- Score decrease > 1.0 → automatic rollback to previous best\n\n---\n\n## Optimization Process\n\nThe optimizer runs up to 3 rounds of evolutionary improvement:\n\n1. **Round 1** — 5 parallel haiku agents generate mutations, fast-rank variants\n2. **Round 2** — Top variants evaluated by sonnet judge panel\n3. **Round 3** — 3×3 parallel sonnet agents for full Layer 2 evaluation of finalists\n\nEach round selects the best-performing variant and uses it as the base for the next iteration.\n\n---\n\n## Integration with Keysarium\n\nBTO works standalone but integrates seamlessly with `@dzhechkov/keysarium`:\n\n```bash\n# Install Keysarium first (optional)\nnpx @dzhechkov/keysarium init\n\n# Then add BTO — it detects Keysarium automatically\nnpx @dzhechkov/skills-bto init\n```\n\nWhen installed alongside Keysarium, BTO can evaluate and optimize any skill or command in the Keysarium toolkit.\n\n---\n\n## Generated skills carry a frontmatter fence\n\nSkills built with `/bto-build` (and the pack's own `SKILL.md`) begin with a `---` frontmatter fence\ncarrying `name:` and `description:` before the `# Title` heading. This is what makes a skill loadable\n— an unfenced `SKILL.md` makes the harness refuse the *entire* skills directory, not just that one\nskill. Commands, rules and agent templates deliberately do **not** get a fence: the parser that\nrequires it reads skills only.\n\nVerify a generated skill with the reproducer the pack itself uses:\n\n```bash\ndz list --skills-dir <project>/.claude/skills   # exit 0 and the skill named in the listing\ndz info <name> --skills-dir <project>/.claude/skills\n```\n\n## Development\n\n```bash\nnpm test    # node --test test/*.test.mjs — pack conformance specs\n```\n\nThe specs assert against the shipped `templates/` text and spawn the real `dz` binary as the\nacceptance oracle. A layout-only check (`dz skills-verify --static`) is necessary but **not**\nsufficient: it passes on a tree whose fence is missing, so `dz list` is the discriminating oracle.\n\n## Requirements\n\n- **Claude Code CLI** — installed and configured ([installation guide](https://docs.anthropic.com/en/docs/claude-code))\n- **Node.js >= 16.0.0** — required for the npm install method\n\n---\n\n## License\n\n[MIT](https://opensource.org/licenses/MIT)\n\n---\n\n## Links\n\n- **GitHub:** [https://github.com/djd1m/dz-harness](https://github.com/djd1m/dz-harness)\n- **Issues:** [https://github.com/djd1m/dz-harness/issues](https://github.com/djd1m/dz-harness/issues)\n- **npm:** [https://www.npmjs.com/package/@dzhechkov/skills-bto](https://www.npmjs.com/package/@dzhechkov/skills-bto)\n- **Keysarium:** [https://www.npmjs.com/package/@dzhechkov/keysarium](https://www.npmjs.com/package/@dzhechkov/keysarium)\n","readmeFilename":"README.md"}