{"_id":"@azure-id/orc","_rev":"16-711c9ed272163e0ea40781461f433a4d","name":"@azure-id/orc","dist-tags":{"latest":"1.8.1"},"versions":{"0.55.0":{"name":"@azure-id/orc","version":"0.55.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@0.55.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"4db1e10a8144e3786631ddc027eebf9fefb9e37f","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-0.55.0.tgz","fileCount":449,"integrity":"sha512-OzshTFzFN3+YgmuGGS5ZgoJi01KP+lslTqM3smFTbZOvIwJoWUc7GYDdgZjUWBFfV/enQ0TzH/id0z6J/3tSkw==","signatures":[{"sig":"MEQCIG8yR0zcTIWCnYEusExbY6qqCA5DzB14lpVPNSbjsCVvAiBHuIEwip02b/CJEaM1mps3I7y4mvFEWparKM7X3Ff/Bw==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":4989964},"engines":{"node":">=18"},"gitHead":"47bfddb16fa26f7ccdbfd2fcea081311c65cfd61","scripts":{"test":"node --test --test-concurrency=6 test/","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node --test --test-concurrency=6 test/","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_0.55.0_1787811202231_0.4975661176544677","host":"s3://npm-registry-packages-npm-production"}},"0.55.2":{"name":"@azure-id/orc","version":"0.55.2","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@0.55.2","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"783fc96d681b43c07d613f37ab57f43cb1943b96","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-0.55.2.tgz","fileCount":449,"integrity":"sha512-JFJGWwQu6ysnyEQy3lc9OB4bUFTTlf6I9KhCAVhy4HL9Gr6j1kX3VKQzG/050JHMisedHz3JmG1JvZp9wLnohQ==","signatures":[{"sig":"MEUCIFIQU5Wted//R92bHjXKfvMHttkiGHzfIIOj0Sx12YECAiEA3LRnJ9TC57mgzidYRM5AXlU+6G7/zp2rP22QKHffaeA=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":4994228},"engines":{"node":">=18"},"gitHead":"e9cfb047cacfd880cf545f9e3f7efbe76fcbce6d","scripts":{"test":"node --test --test-concurrency=6 test/","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node --test --test-concurrency=6 test/","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_0.55.2_1787814898432_0.352094020787858","host":"s3://npm-registry-packages-npm-production"}},"0.56.0":{"name":"@azure-id/orc","version":"0.56.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@0.56.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"8091987a803a7c7d960206e34dd722023e74c147","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-0.56.0.tgz","fileCount":449,"integrity":"sha512-Cw2IaSSzsfWJU+GX3ZKWzPQ8E04moA/8IqD9EXyHUgq304dx/tTM0WdpVGqZdoqhzetfljHefYd2nKqcf19jmA==","signatures":[{"sig":"MEUCIBDFYBl+uhHOdOkdZgwkjZcpI/RTvXitNIQArVsbBVtGAiEA0YqTZChYyBrd7dv+VNr3xahyiXrg+KHL3A6nMQx11es=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":4983459},"engines":{"node":">=18"},"gitHead":"c91d7f01dac33152716772f01d09ef4fe8ad9edd","scripts":{"test":"node --test --test-concurrency=6 test/","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node --test --test-concurrency=6 test/","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_0.56.0_1787834509501_0.9592571174857587","host":"s3://npm-registry-packages-npm-production"}},"0.56.1":{"name":"@azure-id/orc","version":"0.56.1","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@0.56.1","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"5abf71e6e582be5db99077ba9c46cbf4d8bb9ba8","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-0.56.1.tgz","fileCount":449,"integrity":"sha512-niK+4R3ajtVMTKkg2JbEb0ex+RbIWQmP1zRQG6n6CQxxkNeZPEDhQoCnnE+tJJV5VnsyaaLlLpQmS9pEecvr7g==","signatures":[{"sig":"MEQCIG0ZCH0JawRNFw6ysE/ASm+DLEfIkHsQ18k3lilJLuxOAiAr5N349HPUPPa5x3gxrm2eVdkmRYwlNzlZyEM8JrpeBg==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":5075924},"engines":{"node":">=18"},"gitHead":"0ba8382c67f49dbb45c2a406b53834b25b43373c","scripts":{"test":"node --test --test-concurrency=6 test/","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node --test --test-concurrency=6 test/","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_0.56.1_1787898040228_0.8409963114794752","host":"s3://npm-registry-packages-npm-production"}},"1.0.0":{"name":"@azure-id/orc","version":"1.0.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.0.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"e76b472a2f66e5daec3d6b75b3a3420cea6e8390","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.0.0.tgz","fileCount":458,"integrity":"sha512-0VKbajkvhb5A5KKNVHNqAvfIM9KQ/2OjxU3yP8Xi7ZdYvUhGinp1XjaR3mxhMgDe2UBNCr7OtcnPQMNvL6tG4A==","signatures":[{"sig":"MEQCIFl58YqFw8T5n5LRIuVD1LZ/Q5pbfpodR8szotp2HlYcAiAF5SuKmCKkyhCT9uVCkz9mn4pOSweR6P1MlZ7iIlbj1Q==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":5401837},"engines":{"node":">=18"},"gitHead":"f755c149f5871d4bdcb5191a6b872922baf1089d","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.0.0_1788119073696_0.9512013532239532","host":"s3://npm-registry-packages-npm-production"}},"1.1.0":{"name":"@azure-id/orc","version":"1.1.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.1.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"e4598909089644d14713a6d4830a61032ee21375","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.1.0.tgz","fileCount":466,"integrity":"sha512-BUcUJzh0tU1Ktz6pWc6H5zz8M25OLaQub6MYnyTjNHMEzXYPFH0wPo6kEWUydSNw/BVdLpAI5i2qmGtipSKTyQ==","signatures":[{"sig":"MEUCIQDd448CFjzjsMDgGdrAbzFVqBBGvl2icBVhj1hU6qpnEwIgHhNP/1GQIQFg619+6T/FJE+KaZweRmMOnVt8yWW1ejU=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":5502586},"engines":{"node":">=18"},"gitHead":"29b2293dba27f7f7b506a0612276e7d5539444b3","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.1.0_1788196637088_0.9297109251090601","host":"s3://npm-registry-packages-npm-production"}},"1.2.0":{"name":"@azure-id/orc","version":"1.2.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.2.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"b5933376caa471a02543738b29e3dfbc8b35011b","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.2.0.tgz","fileCount":466,"integrity":"sha512-X9wrxDyoaMDGJCOo52QINo/pusz+AamF2kl3OWbNTVEGRtn17pJ0SjGo3cDDWoip9VQ0Bw8T06XYR1pdWG5PuQ==","signatures":[{"sig":"MEQCIELy64i+6+iI84DGAtW8eTYv0L4h7PeUYYBqxQIan0H2AiB3GyrwxDQjXrRYgygBsDqGXI/8nRHPHZuyVXe3ca7eOw==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":5536368},"engines":{"node":">=18"},"gitHead":"cb903042ccdd26b91febe17040f0f083df64b070","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.2.0_1788488301512_0.3814964570082162","host":"s3://npm-registry-packages-npm-production"}},"1.2.1":{"name":"@azure-id/orc","version":"1.2.1","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.2.1","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"c3013fbf6f8d2f46a84d2410cc3c42ea87ae3833","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.2.1.tgz","fileCount":467,"integrity":"sha512-JDlfWGgU7qMTrXYmCQ6KCxSL0AGn3kDawi5JiFHXLu/tEztDk4bBUAFpDeCCVjm7675CWc1pSCqfZeJUn0/FFA==","signatures":[{"sig":"MEYCIQD2idC+jvnO9wvfXcNo/UTHCrYeoiYwnni+RPezHykrNwIhAJQDhTugzXJq3SNmddSASTO87VEJpLQeT7ZlVmlm0SqZ","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":5575387},"engines":{"node":">=18"},"gitHead":"40e897731fbc30fd3cb002a543d52e97099fa887","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.2.1_1788495748840_0.4228477167139588","host":"s3://npm-registry-packages-npm-production"}},"1.4.0":{"name":"@azure-id/orc","version":"1.4.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.4.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"25063b716fc653c1b5784e15ab87f3db155fb488","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.4.0.tgz","fileCount":474,"integrity":"sha512-6WhrVSq7mFoQ5BdyfrfTRqvd6BQ7Du7pccgGs8M0MFHehmErzjSXE6CjuvKuOtli9qHpE2Dit2+ciQ6Mx0TG/g==","signatures":[{"sig":"MEUCICo06t2MAZ1ZfcDl/Uywc1eKWoiVJzP/akOt1FeU9tmuAiEA6BhyryyH1IL+CbqleXEY+jHzxG8AoAz90Fa7lahZ814=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":5913335},"engines":{"node":">=18"},"gitHead":"9e02895dda6588354b25891da99ca1ebd47d5215","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.4.0_1788539101654_0.38149861809644814","host":"s3://npm-registry-packages-npm-production"}},"1.4.1":{"name":"@azure-id/orc","version":"1.4.1","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.4.1","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"a83e7eb92eeabbfcfe11823f24ab8bf49e41677c","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.4.1.tgz","fileCount":474,"integrity":"sha512-rAW8Z/a1feEPZpkQxWWIqOVZz0L4UlNLty7xI/uzGjyfin5saWvUfM60fMjpeJYB7gfRfCbHKivR7y74KdpbWg==","signatures":[{"sig":"MEYCIQDReqOSk6fZbgJ+VsHThEGOMdSpAliq6JJWZWtK6DUYsQIhAL6U0DeG0Bmfl7XSO+8TbHND/psgHV9k0hfQA5wgz4OD","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":6006300},"engines":{"node":">=18"},"gitHead":"638346509aa45b9475ddb8c25cd208b54f0adf43","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.4.1_1788550426939_0.878504429517464","host":"s3://npm-registry-packages-npm-production"}},"1.4.2":{"name":"@azure-id/orc","version":"1.4.2","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.4.2","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"9f438061da6211586d51b171f9f79ea9b18ab03b","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.4.2.tgz","fileCount":474,"integrity":"sha512-OhROPU5or0ka5c5LQepZF/PDohle1D+cGw8NuFhka2YvrWH3w1wHYWEqiTKXqX2RZU5KofoB8TJIGIjKIwT+qA==","signatures":[{"sig":"MEUCIQC67vcGuBTT5i4OzBl9NOvk1cwqLuSt8smk+SAJ+Ssh4wIgHxtsKQ4mo2A4RzobK6aNsN2EqVGNuM6ymWq+U6MqmHk=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":6010860},"engines":{"node":">=18"},"gitHead":"983697da670a313925aff2fbb2d3dd929bf9380a","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.4.2_1788592855029_0.9816723422016669","host":"s3://npm-registry-packages-npm-production"}},"1.5.0":{"name":"@azure-id/orc","version":"1.5.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.5.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"b9aa95fd15b3ffe795efe9011a6fadeb357cbaeb","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.5.0.tgz","fileCount":485,"integrity":"sha512-gBl2b3clUpK44mAbmohUM0Vevrc5Yk/RZBqVA4ocdjuCwKHoT8W2g+O/45qHuVeqC2mKkPqJ+GjCvF19CMeJFw==","signatures":[{"sig":"MEQCIHEV8iAOh+PwJmweHP1d3bF86ncJ1JTKiV9+yKlm5GYaAiA6p7dttxb+pqqy786XJKuYka8I2fWH3XTDVgmGEiB22w==","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":6459828},"engines":{"node":">=18"},"gitHead":"fa154937226522b648dc6c8ff9fc003c790a4787","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.5.0_1788762138382_0.6766766861196805","host":"s3://npm-registry-packages-npm-production"}},"1.6.0":{"name":"@azure-id/orc","version":"1.6.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.6.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"829ad1090885f660996a5dbfd826b2174e53222a","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.6.0.tgz","fileCount":486,"integrity":"sha512-Ra9twmY9oP3SVAp57RJMYC7BGaPHIJN9n4NIqDtfrvB4TZIe3XQnbVom18sc5sstFdSIG1P/KAwnTl6O4KC2kw==","signatures":[{"sig":"MEUCIF6EIV+wGFTvA828XHzUfhNtjCr1xbEy9O/9LCSwTjijAiEAvxps7dD2cwQweUxVjbU7mRhj0ppqfM8QeIAk8OpFHTE=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":6488424},"engines":{"node":">=18"},"gitHead":"7b8e03b46df251081a084ed190d4f2279fb1d14f","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.6.0_1788809559411_0.9107144282149253","host":"s3://npm-registry-packages-npm-production"}},"1.7.0":{"name":"@azure-id/orc","version":"1.7.0","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.7.0","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"06c2250d705ede341fd53636140b658a69dd9cc2","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.7.0.tgz","fileCount":501,"integrity":"sha512-ghZJsc8fJmk/wGn1gY9o9napYf2ZJ0cswU1ReMciNqfxCYLlsPOIWACwfowG80PSnF825LvfEcqh8SXFzJuc5g==","signatures":[{"sig":"MEUCIFHI4MLZsMv6g6J8xGTTgHciStCJagZ6wvx9VXxb2ZjzAiEArmBqj987uYe79jv8Uc8/3j3EhXA6dnWHfN88eT/bJAM=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":6761435},"engines":{"node":">=18"},"gitHead":"7614e3c1fa5207d267f032355037d315b88bd282","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.7.0_1789319862804_0.4667449458318593","host":"s3://npm-registry-packages-npm-production"}},"1.7.1":{"name":"@azure-id/orc","version":"1.7.1","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"license":"MIT","_id":"@azure-id/orc@1.7.1","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"homepage":"https://github.com/azure-id/orc#readme","bugs":{"url":"https://github.com/azure-id/orc/issues"},"bin":{"orc":"bin/cli.js"},"dist":{"shasum":"7dffc45c3ff494129faea01fd4248422bcf56f99","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.7.1.tgz","fileCount":501,"integrity":"sha512-5cCwWSuhrdQ6Hv1+PTjqrgXwEXeCNgTKpollnuOAr9SZ3mVjGN2feGq1ulO9mcbwX4bp5rRj64cbevxGFCQLlg==","signatures":[{"sig":"MEUCIQD3jkaXuMXHtT3PibTcjqf0RXrdEZi6MZG0IWUaUwT5CgIgXTd5mZV2v4fVccbI18sLfkqkzTxJWoul5oH8dnUUeYg=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":6768387},"engines":{"node":">=18"},"gitHead":"97dd81517a20cc82db46fa314298b0fba0093073","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/orc_1.7.1_1789400249880_0.3176013869931953","host":"s3://npm-registry-packages-npm-production"}},"1.8.1":{"_id":"@azure-id/orc@1.8.1","bin":{"orc":"bin/cli.js"},"bugs":{"url":"https://github.com/azure-id/orc/issues"},"dist":{"shasum":"e660fbeaa1fad6bd3d2f1d177dbc2a14a66c960a","tarball":"https://registry.npmjs.org/@azure-id/orc/-/orc-1.8.1.tgz","fileCount":510,"integrity":"sha512-mHyDvnVT0x0btmzNTphQo4BroEj604STH7uSRlaGjmZgLERNFPoeS5l5WOqqVthlzelOOeGIb+KgRrr0k8viOg==","signatures":[{"sig":"MEUCIQDBUfDa17aS4T3S04ofppgo7gY84/HPeh/WhzTFNnsNgQIgWmu2w55GyaGJD2p5hq/RmGrjWBEuS0Eq/Wk+HswEHkE=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"},{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEQCIEM9rFP+8PktKFKBVTmMdc5P9RRoNBWch0N0nS7SG4SaAiABgNEHBmXvlhSw4102IhYv5v9etaJDTjNpzC0PmP3JkQ=="}],"unpackedSize":7074562},"name":"@azure-id/orc","engines":{"node":">=18"},"gitHead":"bc41f26db25cf8a205e1d36b7ca157f0fe21dabb","license":"MIT","scripts":{"test":"node bin/test-run.js","verify":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js","prepack":"node bin/build-agents.js --check && node bin/verify-package.js && node bin/verify-contracts.js && node bin/test-run.js","postinstall":"node -e \"try{require('fs').accessSync(require('path').join(__dirname,'bin','cli.js'));console.log('\\nORC installed. Run: orc init (or orc init --global)\\n')}catch(e){console.error('\\n[orc] WARNING: bin/cli.js missing from this install — the repo/publish was incomplete. Reinstall from a complete source.\\n')}\"","build:agents":"node bin/build-agents.js"},"version":"1.8.1","_npmUser":{"name":"azure-id","email":"azuredev.id@gmail.com"},"homepage":"https://github.com/azure-id/orc#readme","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"_npmVersion":"10.8.2","description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","directories":{},"maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"_nodeVersion":"20.17.0","_hasShrinkwrap":false,"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/orc_1.8.1_1789526262068_0.6844094300372161"}}},"time":{"created":"2026-08-27T06:13:22.038Z","modified":"2026-09-16T02:37:42.454Z","0.55.0":"2026-08-27T06:13:22.414Z","0.55.2":"2026-08-27T07:14:58.637Z","0.56.0":"2026-08-27T12:41:49.684Z","0.56.1":"2026-08-28T06:20:40.384Z","1.0.0":"2026-08-30T19:44:33.921Z","1.1.0":"2026-08-31T17:17:17.257Z","1.2.0":"2026-09-04T02:18:21.757Z","1.2.1":"2026-09-04T04:22:29.021Z","1.4.0":"2026-09-04T16:25:01.837Z","1.4.1":"2026-09-04T19:33:47.143Z","1.4.2":"2026-09-05T07:20:55.182Z","1.5.0":"2026-09-07T06:22:18.554Z","1.6.0":"2026-09-07T19:32:39.647Z","1.7.0":"2026-09-13T17:17:43.004Z","1.7.1":"2026-09-14T15:37:30.072Z","1.8.1":"2026-09-16T02:37:42.268Z"},"bugs":{"url":"https://github.com/azure-id/orc/issues"},"license":"MIT","homepage":"https://github.com/azure-id/orc#readme","keywords":["claude","claude-code","orchestrator","skills","agents","ai"],"repository":{"url":"git+https://github.com/azure-id/orc.git","type":"git"},"description":"ORC — an orchestrator skill constellation for Claude Code: intake, planning, scored parallel subagents, code-pattern matching, review, verify, ship, plus a project knowledge-base wiki.","maintainers":[{"name":"azure-id","email":"azuredev.id@gmail.com"}],"readme":"
\r\n\r\n# 🐋 ORC\r\n\r\n**An orchestrator skill constellation for [Claude Code](https://claude.com/claude-code).**\r\n\r\n*Intake → analyze → plan → score → parallel subagents → review → verify → ship.*\r\n\r\n![npm](https://img.shields.io/npm/v/%40azure-id%2Forc?style=for-the-badge&color=cb3837&logo=npm)\r\n![Version](https://img.shields.io/badge/version-1.8.1-blue.svg?style=for-the-badge)\r\n![License](https://img.shields.io/badge/license-MIT-green.svg?style=for-the-badge)\r\n![Node](https://img.shields.io/badge/node-%3E%3D18-brightgreen.svg?style=for-the-badge)\r\n![Claude Code](https://img.shields.io/badge/Claude_Code-Skills-purple.svg?style=for-the-badge)\r\n![Dependencies](https://img.shields.io/badge/dependencies-zero-lightgrey.svg?style=for-the-badge)\r\n![GitHub stars](https://img.shields.io/github/stars/azure-id/orc?style=for-the-badge&color=yellow)\r\n\r\n**Latest: v1.8.1** · updated 2026-09-16 · [full changelog](CHANGELOG.md)\r\n\r\n**On npm: [`@azure-id/orc`](https://www.npmjs.com/package/@azure-id/orc)** — `npm i -g @azure-id/orc`\r\n\r\n**🇮🇩 [Baca dalam Bahasa Indonesia](README-id.md)**\r\n\r\n
\r\n\r\n---\r\n\r\n> [!CAUTION]\r\n> **Upgrading from a version before v0.56.0? Do this once.**\r\n>\r\n> The package moved from the unscoped `orc` to **`@azure-id/orc`**. Both declare\r\n> the same `orc` command, and npm will not hand that command to the new package\r\n> while the old one still holds it — so **every** install source fails with the\r\n> same error, and `orc upgrade` cannot fix itself:\r\n>\r\n> ```text\r\n> npm error code EEXIST\r\n> npm error File exists: C:\\Users\\you\\AppData\\Roaming\\npm\\orc\r\n> ```\r\n>\r\n> Run these two lines once. Nothing in your `.claude/` is touched, and your\r\n> `orc.config.yaml` survives:\r\n>\r\n> ```bash\r\n> npm uninstall -g orc # release the `orc` command from the old package\r\n> npm i -g @azure-id/orc # install the current one\r\n> orc update # re-apply into this project (add --global for ~/.claude)\r\n> ```\r\n>\r\n> **From v0.56.0 onward `orc upgrade` handles this for you** — it removes the old\r\n> package first, then installs, and says so while it does it. `orc doctor` also\r\n> reports the old package by name if it is still there.\r\n>\r\n> Do **not** reach for `npm i -g -f`. `--force` overwrites the command file and\r\n> leaves the superseded package installed underneath, owning nothing and never\r\n> updated again.\r\n\r\n---\r\n\r\n> [!IMPORTANT]\r\n> **`orc-open` is released — ORC for non-Claude agents.**\r\n> A provider-agnostic port of the pipeline:\r\n> **[github.com/azure-id/orc-open](https://github.com/azure-id/orc-open)**. Use it\r\n> if you run a different coding agent. This repository stays Claude Code–native.\r\n\r\n---\r\n\r\n## What ORC is\r\n\r\nYou give ORC a feature — or a requirements document. It works out what you\r\nmeant, plans the work, sends each task to the **cheapest model that can still do\r\nit**, runs the tasks that do not collide **at the same time**, reviews the\r\nresult, verifies it against a definition of done you signed off, and ships it.\r\n\r\nIt writes its state to disk as it goes, so a long run survives a pause, a token\r\nlimit, or a brand new chat session.\r\n\r\n**ORC is not a program that runs.** It is a set of markdown **skills**, **slash\r\ncommands** and **subagent definitions** that Claude Code reads and follows. This\r\nzero-dependency npm package copies those files into your `.claude/` folder.\r\n\r\n```text\r\n ┌──────────────── you own scope + sign-off ────────────────┐\r\n feature / doc ──▶ intake ─▶ analyze ─▶ plan ─▶ score ─▶ ⇉ parallel waves ⇉ ─▶ review ─▶ verify ─▶ ship\r\n (grounded) (per task) (cheapest capable model) (checkpointed to disk)\r\n```\r\n\r\n---\r\n\r\n## 👀 See it run before you run it\r\n\r\nEvery lane is written out as a **mocked run**: what you type, what ORC prints\r\nback, and the files that land on disk. Nothing was executed to make them — they\r\nexist so you never have to spend tokens to find out what a command does.\r\n\r\n### **▶ [Start here: `mock-run/INDEX.md`](mock-run/INDEX.md)**\r\n\r\nAlso available without leaving your machine:\r\n\r\n```bash\r\norc mock-run list # every walkthrough, in reading order\r\norc mock-run show orc-pact # read one\r\norc ui # ▸ Mocked Skill Use — the same docs, searchable\r\n```\r\n\r\n---\r\n\r\n## Why it works this way\r\n\r\nA single agent handed a real feature fails the same ways every time: it silently\r\npicks one reading of your request, runs the most expensive model on everything,\r\nforgets decisions when the context compacts, says \"done\" against a definition\r\nnobody wrote, cites code that does not exist, and leaves nothing to inspect.\r\n\r\nThose are **process problems** — the ones teams solved with roles, reviews and\r\nwritten agreements. ORC writes that discipline down as skills:\r\n\r\n- **Coordinating and doing are different jobs.** The orchestrator never\r\n implements. Even a one-line change goes to a subagent, which keeps its own\r\n context lean for the whole run.\r\n- **Every task is scored, and the score picks the model.** You see the table\r\n before anything starts, and the agents are named and model-pinned, so what ran\r\n is a fact you can check.\r\n- **\"Done\" is written before the work starts.** Intake produces a signed-off\r\n spec whose definition of done becomes the final verification.\r\n- **Nothing is trusted, everything is attested.** `file:line` quotes, verbatim\r\n build output, anchored findings — and the orchestrator spot-checks them, so a\r\n made-up citation bounces instead of riding into a task.\r\n- **Disk beats memory.** Eager checkpoints turn every pause into a clean resume,\r\n including in a fresh session.\r\n- **Rigor is a dial.** The same spine runs as `/orc-mini` (one subagent), `/orc`\r\n (real features), and `/orc-ultra` (advisor plus judgment gates).\r\n- **It learns.** Code patterns make executors write like your codebase, the wiki\r\n sharpens every future plan, and traces feed `/orc-retro`, which recalibrates\r\n the scoring from real runs.\r\n\r\n---\r\n\r\n## Quick start\r\n\r\nORC is on npm as **[`@azure-id/orc`](https://www.npmjs.com/package/@azure-id/orc)**.\r\n\r\n```bash\r\nnpm i -g @azure-id/orc # install\r\nnpm i -g @azure-id/orc@latest # update to the newest release\r\n```\r\n\r\n
\r\nInstall straight from GitHub instead\r\n\r\n
\r\n\r\n```bash\r\nnpm i -g https://github.com/azure-id/orc/archive/refs/heads/main.tar.gz\r\n```\r\n\r\n
\r\n\r\nThen, inside a project:\r\n\r\n```bash\r\norc init # install into ./.claude (this project)\r\norc init --global # install into ~/.claude (all projects)\r\norc onboarding # the whole walkthrough in the terminal — no GitHub needed\r\norc config # view or change settings (zero model tokens)\r\norc ui # the local control panel\r\norc --help # every command\r\n```\r\n\r\nAfter installing:\r\n\r\n1. Paste your team's PR template into `skills/orc/subskills/orc-pr/pr.md`.\r\n2. Add `.claude/orc/run/` to your project `.gitignore`.\r\n3. Run **`/agents`** to confirm your Claude Code accepts the agent model ids.\r\n4. **Run your main session on Opus 5.** A subagent can never use a better model\r\n than your session. This is the most common cause of \"it used the wrong\r\n model\" — see [model selection](guides/model-selection.md).\r\n5. If a `/command` does not appear, your Claude Code may read commands from a\r\n different folder — move the files in `commands/` there.\r\n\r\n
\r\nStaying up to date\r\n\r\n
\r\n\r\n`orc update` re-copies the files already in this package. It never uses the\r\nnetwork. **`orc upgrade` is the one that pulls a new version**: it fetches the\r\nnewest package first, then applies it. Your `.claude/orc.config.yaml` survives\r\nboth.\r\n\r\n```bash\r\norc version # what you have, and whether something newer exists\r\norc changelog # what you would GET by upgrading\r\norc upgrade # fetch the latest, then update this project\r\norc upgrade --global # the same for ~/.claude\r\norc upgrade --from @azure-id/orc # explicitly from npm\r\norc upgrade --from github:azure-id/orc # a fork, or any npm spec\r\n```\r\n\r\nOr update the package yourself and re-apply it:\r\n\r\n```bash\r\nnpm i -g @azure-id/orc@latest\r\norc update\r\n```\r\n\r\nThe update check reads the source over HTTPS, is cached for 24 hours, and fails\r\nsilently when you are offline. Turn it off with `ORC_NO_UPDATE_CHECK=1`.\r\n\r\nYou do not have to run a command to hear about it: the same notice appears\r\ninside Claude Code through ORC's hooks, at **zero model tokens** — hooks are\r\nscripts Claude Code runs, not model turns.\r\n\r\n`orc upgrade` tries the npm registry first, then a plain tarball, then the\r\nGitHub spec — and it remembers which one worked. If the old unscoped `orc`\r\npackage is still installed, it removes that first (announced), because npm\r\ncannot give the `orc` command to `@azure-id/orc` while another package owns it.\r\nSee the caution at the top of this README for the one-time manual version.\r\n\r\n
\r\n\r\n> **\"ORC cannot see my wiki\"?** Run **`orc wiki sync`**, not a new scan. Docs\r\n> without a manifest are *unregistered*, not missing — common when a scan stopped\r\n> at one of `/orc-wiki`'s pauses. Sync rebuilds the index from the docs you\r\n> already have, for free.\r\n\r\n> **\"What does ORC actually know about my project?\"** `orc wiki docs` lists every\r\n> registered doc, `orc wiki coverage` says how much of your code is written about\r\n> at all, and `orc pattern show ` prints the conventions that go into every\r\n> agent that writes code here. All free, all read-only —\r\n> **[`guides/knowledge-reads.md`](guides/knowledge-reads.md)**.\r\n\r\n---\r\n## Terminal Hook\r\nORC have terminal hook to see: Context Window %, 5 Hour usage %, Weekly usage % and others you might see\r\n\r\n\"image\"\r\n\r\n
\r\n\r\n---\r\n\r\n## The lanes\r\n\r\n> [!TIP]\r\n> They chain naturally: **`/orc-brainstorm` → `/orc-grill` → `/orc-analyze` →\r\n> `/orc-plan` → `/orc-route` → `/orc`**. You can start anywhere.\r\n\r\n### Build something\r\n\r\n| Command | What it does | Mocked run |\r\n|---|---|---|\r\n| **`/orc`** | The full pipeline: intake → plan → scored parallel waves → review → verify → ship. Checkpoints eagerly; resumes in a fresh session. | [see it](mock-run/orc.md) |\r\n| **`/orc-ultra`** | The same, plus an Opus 5 **xhigh** advisor and three judgment gates. Deep analysis, patterns, tests and security forced on. Costly by design. | [see it](mock-run/orc-ultra.md) |\r\n| **`/orc-mini`** | One Sonnet 5 executor, a build + test smoke gate, ship. Skips full review and verify. Switches to the full flow mid-run on request. | [see it](templates/skills/orc-mini/examples/mini-run-mock.md) |\r\n| **`/orc-fast`** | The fastest lane. Needs a fresh wiki **and** a cached code pattern; then it skips the analyst and planner entirely. A missing prerequisite falls back to `/orc-mini` — the chat never stops. | [see it](mock-run/orc-fast.md) |\r\n| **`/orc-quick`** | Ask for anything: a fix, a question, a defect hunt, a dependency bump, PR comments. Look → ask once → do. **It always asks which agent to dispatch**, and no setting can change that. | [see it](mock-run/orc-quick.md) |\r\n| **`/orc-wait`** | Wall-clock pause without losing the run. You see the window is nearly full, type `/orc-wait 30`, and ORC hands the run back to disk, waits in detached hops that **cost zero tokens**, and picks up where it stopped. Three modes decide how much finishes first: `safe` · `soft` (forces the checkpoint) · `hard` (fastest, can lose an in-flight return). `/orc-wait block ` tells it not to stop you at all. | — |\r\n| **`/orc-diy`** | Your own lane, composed in the terminal with `orc diy` and compiled. Unconfigured or stale → it refuses and offers plain `/orc`. | [see it](mock-run/orc-diy.md) |\r\n\r\n### Work out what to build\r\n\r\n| Command | What it does | Mocked run |\r\n|---|---|---|\r\n| **`/orc-brainstorm`** | You do not have the idea yet. It generates candidates against named thinking lenses with **no criticism while generating**, clusters them into a few real directions, stress-tests each, then recommends one and **waits — it never picks for you**. Every menu ends with your own slot. | [see it](mock-run/orc-brainstorm.md) |\r\n| **`/orc-grill`** | You have one idea and it is still vague. It asks rounds of questions, **looks facts up itself** instead of making you recite your own codebase, and never answers its own question. Ends when *you* say the idea matches what you meant. | [see it](mock-run/orc-grill.md) |\r\n| **`/orc-analyze`** | A document or a request → a scope-bounded, code-grounded spec. Every claim carries `file:line` evidence or becomes a question. Deep mode adds parallel scouts. | [see it](templates/skills/orc-analyze/examples/analyze-mock.md) |\r\n| **`/orc-plan`** | A request or a spec → a real task plan: grounded files, dependencies, facets, and a test disposition per task. | [see it](mock-run/orc-plan.md) |\r\n| **`/orc-doc`** | Writes the long document — a PRD, a TSD, a cross-team agreement, a status report or a runbook — as portable Markdown that imports cleanly into Notion, Obsidian, Docs, Coda, Craft and GitHub. **ORC never reads the document body**: each section is its own file under `sections/`, each writer owns exactly one of them, each checker reads one bounded part, and `document.md` is a build artifact rebuilt for free. Every wave is a stop you can walk away from, and it resumes months later without you explaining anything twice. | [see it](mock-run/orc-doc.md) |\r\n| **`/orc-route`** | You have a plan — which lane should build it? It names one lane, the runners-up with what each costs you, and any lane that is impossible with the condition blocking it. **It refuses to route a sentence**, because that would be guessing. | [see it](mock-run/orc-route.md) |\r\n| **`/orc-explain`** | \"Wait, what?\" It says the last message again: the point first, then the background it assumed, then every ORC-only word defined in your project's terms. | [see it](mock-run/orc-explain.md) |\r\n| **`/orc-poly`** | One change across two or more repos, without drift. Peer source is read-only; it freezes the shared boundary into a contract and writes one plan per repo. It never builds. | [see it](templates/skills/orc-poly/examples/poly-run-mock.md) |\r\n\r\n### Teach ORC your project\r\n\r\n| Command | What it does | Mocked run |\r\n|---|---|---|\r\n| **`/orc-wiki`** | Scans your codebase into a persistent `wiki/`, evidence-anchored, and points `CLAUDE.md` at it. Freshness is computed on read, never stored. Expensive and opt-in — it always warns first. | [see it](templates/skills/orc-wiki/examples/wiki-run-mock.md) |\r\n| **`/orc-pattern`** | Learns your real conventions per language, so executors write code that matches your codebase. Your conventions win; security and correctness invariants always carry through. | [see it](mock-run/orc-pattern.md) |\r\n| **`/orc-learn`** | Onboarding docs for a human, one feature at a time, `file:line`-anchored. Local and git-ignored. | [see it](templates/skills/orc-learn/examples/learn-run-mock.md) |\r\n| **`/orc-claude`** | Builds or refreshes this repo's `CLAUDE.md` from verified facts. Zero questions; your own content is never trimmed. | [see it](templates/skills/orc-claude/examples/claude-run-mock.md) |\r\n| **`/orc-export`** | Compiles the wiki, patterns, `PACT.md` and boundary cards into a portable `AGENTS.md` — derived, fingerprinted, `--check`able. So ORC is not a trap. | [see it](mock-run/orc-export.md) |\r\n\r\n### Check what happened\r\n\r\n| Command | What it does | Mocked run |\r\n|---|---|---|\r\n| **`/orc-challenge`** | Grades a **finished** artifact — a TSD, a PRD, an ADR, an API contract, a README, a module — against a goal **you** state, then **stops and makes you fix it in a different session**. ORC judges, you fix, ORC re-judges: **it never fixes what it judged**, because a session that just wrote the fix would grade its own homework. **And it never guesses what \"good\" means here.** | [see it](mock-run/orc-challenge.md) |\r\n| **`/orc-pact`** | The promises your project makes, and which are in doubt right now. Four states, all **computed on read**: HOLDING · DRIFTED · **UNCHECKABLE** (the honest state — never a failure) · BROKEN. It never invents a promise and never retires one for you. | [see it](mock-run/orc-pact.md) |\r\n| **`/orc-boundary`** | What ORC should **not** try here, and exactly what would change that. EXECUTE · ESCALATE · REFUSE, per area. **A REFUSE always names what would make it a yes.** It gates ORC's own dispatch, never your instructions. | [see it](mock-run/orc-boundary.md) |\r\n| **`/orc-test`** | **Runs** the test against a running system - yours, or one you were given permission to touch. Sends the requests, writes down what came back, and stops. **The CLI executes and measures; the model designs and interprets.** Three verdicts, and **`unknown` is the honest one**. The OWASP API Top 10 (2023) as a **closed set** where a category it could not measure keeps its slot and never becomes a pass. It never edits the system it is testing. | [see it](mock-run/orc-test.md) |\r\n| **`/orc-verify`** | Verifies only your git-modified changes: build, tests, diff sanity, findings on a P0–P3 ladder. Read-only. | [see it](templates/skills/orc-verify/examples/verify-mock.md) |\r\n| **`/orc-aftermath`** | Did what we shipped hold up? Graded from the repository's own future: churn, reverts, deleted tests, broken promises. **Churn is a signal, never a verdict**, and it never names a person. | [see it](mock-run/orc-aftermath.md) |\r\n| **`/orc-budget`** | What a run costs, in the unit you are billed in. A **token vector** — fresh input, cache write, cache read, output, never blended — shown four ways: tokens, dollars, percent of your 5-hour window, and context risk. Needs a plan, not a sentence. | [see it](mock-run/orc-budget.md) |\r\n| **`/orc-retro`** | Mines the behavior traces into a calibration report and files it upstream as a PR. | [see it](templates/skills/orc-retro/examples/retro-mock.md) |\r\n\r\n### Ship and hand over\r\n\r\n| Command | What it does | Mocked run |\r\n|---|---|---|\r\n| **`/orc-pr-setup`** | Decides where a big change gets cut into stacked pull requests: ordered layers, each with a purpose, a file list and a measured budget. It stops and asks at every uncertain seam, and never touches git. | [see it](mock-run/orc-pr-setup.md) |\r\n| **`/orc-pr-driver`** | Executes that plan: a branch per layer, a **mandatory green gate at each layer's own base**, `gh stack submit`, then restack and bottom-up merge. | [see it](mock-run/orc-pr-setup.md) |\r\n| **`/orc-handoff`** | For someone who does not read code. The grade comes from **whether a cheap check exists**, not from the file type. It shows the undo command *before* it writes, and never touches a red file. | [see it](mock-run/orc-handoff.md) |\r\n\r\n---\r\n\r\n## Rules that keep the slop out\r\n\r\nORC writes a lot of prose and a lot of code. Both come out carrying the same\r\nrecognisable defaults: sentences that say nothing in a confident shape, and code\r\nthat is longer, more defensive and more abstract than the task asked for.\r\n`orc rules` is the filter, and it has **two halves that never mix**.\r\n\r\n| Half | Who writes it | Where | Changes when |\r\n|---|---|---|---|\r\n| **ORC rules** | ORC | `.claude/skills/_shared/rules/` | `orc update` |\r\n| **Your rules** | you | `.claude/orc/rules.md` | you type |\r\n\r\n**65 shipped rules, read-only**, in four packs: `OSW` writing (23) · `OSC` code\r\n(22) · `OSD` delivery (10), which is what an agent reports about its OWN work ·\r\n`OSU` UI (10), which rides per TASK rather than per lane. Three tiers - **HARD**\r\nabsolute, **PURPOSE** allowed with a written one-line reason, **LOCK** a\r\nconsistency check.\r\n\r\n**Your rules are plain text**, three headings, as much under each as you want -\r\nand **they beat an ORC rule outright**:\r\n\r\n```\r\nhouse rules > your rules > ORC rules\r\n```\r\n\r\nThe house rules are about CODE and behaviour. They say nothing about the words\r\nan agent writes, so they never overrule a writing rule. To switch an ORC rule\r\noff, name its id in a rule of your own; the override is then counted, printed at\r\npreflight, and stated inside every slice. **It is never silent.**\r\n\r\n```bash\r\norc rules # both halves, the ladder, the counts\r\norc rules add --priority P0 --text \"...\" # your own, in your own words\r\norc rules lint src/ README.md # free, zero tokens\r\norc rules credits # every source, author and licence\r\n```\r\n\r\n`orc rules lint` checks **13 of the 65** - the ones a string match can prove -\r\nand prints, every time, `not checked here: 52 rules. They need a reader, not a\r\nmatcher.` A clean lint is not a clean review, and it says so.\r\n\r\n**The rules are other people's work, and the credit ships with them.** Adapted\r\nfrom [`petergyang/no-ai-slop`](https://github.com/petergyang/no-ai-slop) (MIT),\r\n[`miqdadbadjuber/anti-slop`](https://github.com/miqdadbadjuber/anti-slop) (MIT),\r\n[`ehmo/slopkit`](https://github.com/ehmo/slopkit),\r\n[`BioInfo/slopless`](https://github.com/BioInfo/slopless), Andrej Karpathy's\r\n`CLAUDE.md`, Matty Cartwright's Anti-Slop Writing Rules, and three papers on LLM\r\ncode smells.\r\n\r\n**Full detail: [`guides/rules.md`](guides/rules.md).**\r\n\r\n---\r\n\r\n## Documents that go somewhere\r\n\r\n`/orc-doc` writes the long document — and Markdown is the deliverable because of\r\nwhere a Markdown file can actually go:\r\n\r\n| Target | Imports `.md`? |\r\n|---|---|\r\n| Notion · Obsidian · Google Docs · Coda · Craft · Apple Notes · GitHub | **natively** |\r\n| Docusaurus · Hugo · Jekyll · MkDocs | yes — and these *want* YAML front matter |\r\n| Confluence | not natively. Plan for a marketplace importer app |\r\n| Microsoft OneNote | **no**. Convert to Word or PDF first |\r\n\r\nThat table is load-bearing, not decoration: `orc doc lint --target` enforces the\r\nreal limit of the place your document is going. Notion has three heading levels,\r\nso an H4 is an **error** there. A hard-wrapped paragraph is an error everywhere,\r\nbecause a wrap at 80 columns becomes a line break inside a Notion paragraph.\r\n\r\nFive base templates — `prd` · `tsd` · `collaboration` · `report` · `workflow` —\r\neach a floor rather than a cage. `orc doc templates` prints them; bring your own\r\nand its headings become the outline.\r\n\r\n**Full detail: [`guides/documents.md`](guides/documents.md).**\r\n\r\n---\r\n\r\n## The code graph — `orc graph`\r\n\r\nA local map of how your code is connected. It is **off by default**.\r\n\r\n```bash\r\norc config set code_graph on # code lanes build and use the map (free)\r\norc config set code_graph_notes wave # optional: one-sentence notes (costs tokens)\r\norc graph ctx OrderService.create # a card: callers, calls, SQL/HTTP/env effects\r\norc graph changes # what THIS diff touched, and how risky\r\norc graph cochange src/orders.js # what usually changes with this file\r\norc graph coverage src/orders.js # how much of it the parser really saw\r\n```\r\n\r\n- **Structure is free.** The CLI parses the code. No model, no dependency.\r\n Functions, methods, classes and route handlers (`GET /orders/:id`); a\r\n middleware passed by name is a `used by` link.\r\n- **Updates are small.** Git already hashes every file. Only the files whose hash\r\n changed are parsed again — a teammate's change heals at the next preflight.\r\n- **Every link says how sure it is:** `LOCAL`, `IMPORT`, `UNIQUE`, `AMBIGUOUS`\r\n (every candidate listed) or `UNRESOLVED`. The graph never guesses.\r\n- **A card has a token budget.** It never goes over it, and it says what it hid.\r\n `--format tree` names each column once instead of on every row, and falls back\r\n to the normal shape when a card is too small to pay for the header.\r\n- **It shows where code is. It does not replace reading it.** Agents still read\r\n the line range before they act. A card header says `coverage partial 327-466`\r\n when the parser could not finish a file — **no recorded gap is not proof of\r\n completeness**. A card also lists only the callers that NAME the symbol: one\r\n that reaches it through an HTTP route or a string dispatch is not a link, so\r\n **a card's silence is not proof of absence**.\r\n- **Every answer carries a `generation`** — a number that goes up each time the\r\n map changes. A card quoted back later can be placed in time.\r\n- **It keeps itself current in three ways, and only one of them needs a lane to\r\n remember a step.** A read repairs the files it is about to answer for; the\r\n installed hook updates the map when a worker finishes; and every code lane\r\n — `/orc`, `/orc-ultra`, `/orc-diy`, `/orc-mini`, `/orc-fast`, `/orc-quick` —\r\n still builds it at preflight and updates it after each change. Other lanes\r\n never call it. **Nothing runs on a timer and nothing runs in the background.**\r\n- **The hook also hands a worker the anchors it would otherwise search for**, and\r\n everything it hands over is labelled repository data, never an instruction.\r\n Switch it off with `orc config set code_graph_hooks off`; the map still works.\r\n\r\nThe contract: `templates/skills/_shared/code-graph.md`. The hook:\r\n`templates/hooks/README.md`.\r\n\r\n> **It does not save tokens, and we measured that rather than guessing.** Across\r\n> 242 real ORC lane windows, `Grep` and `Glob` results are **0.06%** of what a\r\n> session adds to its context and every tool result together is 5.5%. A perfect\r\n> locator — every whole-file read turned into a range read — would save **0.1% of\r\n> a session**. Turn the map on because you want the cards: what calls what, what\r\n> a change would touch, where the parser could not finish. Not for a number.\r\n> The working is in `eval/graph-replay.js`, and it spends no tokens to re-run.\r\n\r\n## `orc ui` — the control panel\r\n\r\nA local web page for **everything in ORC that is not ai**. One boundary defines\r\nit: **it never runs a lane, never spawns `claude`, never calls a model API.**\r\nEverything it shows or writes is deterministic CLI output.\r\n\r\n\"image\"\r\n\r\n> 🎬 **Video walkthrough — not recorded yet.** The player belongs here; see\r\n> [`mock-run/media/README.md`](mock-run/media/README.md) for the two files to\r\n> drop in. Until then, the panel is written out screen by screen in\r\n> [`mock-run/orc-ui.md`](mock-run/orc-ui.md).\r\n\r\n```bash\r\norc ui # binds 127.0.0.1:9921 and opens a browser\r\norc ui --port 9930 # an explicit port never auto-walks — a collision is an error\r\norc ui --no-open # print the URL only\r\norc ui --idle 0 # disable the idle shutdown (default: 30 minutes)\r\norc ui --fixtures # canned data, no project needed\r\norc ui --stop # shut this project's server down\r\n```\r\n\r\n| Panel | Shows | Can change |\r\n|---|---|---|\r\n| Overview | version, `orc doctor`, wiki tier, what is waiting — plus **Worth doing**, one list of everything wanting a decision | — |\r\n| Settings | every config key, grouped, each with its own control | staged edits, applied together |\r\n| Runs | run history as an accordion: a row opens in place into state-of-play, resume prompt, checkpoint, trace tail | — |\r\n| **Knowledge** | **five tabs**: the wiki's tier AND its **contents** (every doc, what it covers, how often it is read), coverage against your tracked files, the code patterns with the conflicts the codifier flagged, repair memory with a **preview-then-apply** prune, and a read-only view of the linked repos | `wiki sync`, `gotcha prune` |\r\n| Stats | lane and agent usage, downgrades, and a **Cost** tab whose stacked bar keeps cache-read visible | — |\r\n| Flow | the compiled DIY flow, its gate, and a stepper of every phase in order | `diy set`, `diy compile`, presets |\r\n| Crosslink | **Design** (the boundary as a graph) and **Settings** (each peer's freshness) | `crosslink add` / `remove` |\r\n| Promises · Boundary · Self-serve | the pact ledger, the boundary cards, and the surfaces a non-developer can change | `pact check`, `pact sync`, `handoff set` |\r\n| **Docs** | every `/orc-doc` document as a **ribbon** — one block per section, sized by its length and coloured by its state — plus the section files with their sub-parts, the wave strip, the lint health card and the wave preview | `doc compile` · `doc migrate` |\r\n| **Extra** | **six tabs**: the connection setup, the local tools and their state, the band ladder and the six positions, spending per profile per band, and **Recovery** — every dispatch that never reported back, with what it left on disk | `extra add` / `ping` / `route` / `role` |\r\n| **Challenge** | every `/orc-challenge` cycle: the frozen goal, the council roster and what each lens raised, the findings with their dispositions, and whether the pass is computed or blocked | `challenge record` · `accept` · `rebut` |\r\n| **Mocked Skill Use** | every mocked run that ships with ORC, grouped and searchable, with a reading pane | — |\r\n| Learn | the `orc onboarding` walkthrough, one section at a time | — |\r\n| Experiment | every lane with a copy button; opens a Claude session in a terminal | — |\r\n| Maintenance | `update`, `update --prune`, `doctor --fix`, `upgrade` | preview, then apply |\r\n\r\n- **The panel *is* the CLI.** It reads `orc --json` and shells the real\r\n command for every write, so it cannot drift from the CLI — it has no second\r\n copy of anything.\r\n- **A free action gets a button. A paid action gets a command to copy.**\r\n- **Nothing is automatic**, and a prune names **every** file. A count is not\r\n consent.\r\n- **Treated as a write surface**: loopback only, a fresh token per launch, a\r\n Host-header check against DNS rebinding, no CORS, POST-only mutations.\r\n- **Project-scoped, no `--global` config.** If a global install exists that could\r\n win skill resolution, every page carries a banner. It reports that; it never\r\n edits global config.\r\n- **English and Indonesian.** Only the panel's own words are translated — config\r\n keys, model ids, paths, commands and doctor messages are printed exactly as the\r\n CLI wrote them, because a translated config key is a key that does not exist.\r\n\r\nZero dependencies, zero build step: `node:http`, plain JavaScript, hand-written\r\nCSS.\r\n\r\n---\r\n\r\n## Running part of the ladder somewhere else — `orc extra`\r\n\r\n**The orchestrator is always Claude.** What Extra changes is *who executes a\r\nslice*: a score band you own, or one of six named positions, can point at\r\nDeepSeek, GLM, Kimi, MiniMax, Qwen, a local Ollama, or a coding CLI you already\r\nhave signed in (opencode, codex). Everything downstream — the smoke gate, the TDD\r\ngate, the reviewer, the worktree-delta check — is engine-blind, so nothing learns\r\nthe work was foreign.\r\n\r\n**Off by default, and it cannot be armed until something has actually answered.**\r\n\r\n```bash\r\norc extra providers # the shipped, dated catalog — providers, never models\r\norc extra tools # local CLIs: absent · outdated · unauthenticated · ready\r\norc extra add ds --provider deepseek --engine api --env-key DEEPSEEK_API_KEY\r\norc extra ping ds # the connection gate: a ladder, and nothing reads stronger than it is\r\norc extra models ds --test # a LISTED model can still be dead upstream\r\norc extra health ds --model # …and a working model is not one that FINISHES\r\norc extra route set 40-55 ds/deepseek-chat # a score band\r\norc extra role set doc-writer ds/deepseek-chat # or a named position\r\norc config set extra_enabled true\r\n```\r\n\r\n- **The catalog ships providers and never models.** A shipped model id is wrong\r\n within a quarter and wrong *silently* — a 404 mid-wave. `orc extra ping` reads\r\n the live list and caches it; nothing invents a name. Same for price: a cost\r\n figure ORC did not price itself is never printed, it reads as an em dash.\r\n- **Every armed run says so before wave 1.** Routing work off Claude silently is\r\n the failure this whole subsystem is shaped around.\r\n- **Two hard hold-backs**: a task with a cited `risk[]` (auth, money, migration,\r\n security, concurrency, data-integrity) stays on Claude unless you say\r\n otherwise, and a `/orc-boundary` REFUSE area holds even in `warn`.\r\n- **A foreign return is foreign input.** It is the only foreign class that edits\r\n your worktree, so what it says it did is a *claim*, checked against the tree.\r\n- **Your key never reaches a command line.** It travels on stdin into an\r\n encrypted vault, or it stays in an environment variable, or the tool holds its\r\n own — and the passphrase is a **deadline**, not a second factor.\r\n- **Six positions for the lanes that pin an agent instead of scoring a task**:\r\n `quick-executor` · `fast-executor` · `doc-writer` · `doc-checker` ·\r\n `wiki-scanner-deep` · `wiki-scanner-light`. A position with no row keeps its\r\n slot and reads as its pinned Claude agent — \"I left the checker on Claude on\r\n purpose\" and \"there is no checker\" must never look the same.\r\n\r\n**When a foreign worker fails, it is a position and not a blank page.** ORC\r\njournals the baseline *before the first byte leaves the machine*, so a worker cut\r\noff mid-write is **reconciled and resumed** — never re-dispatched from scratch\r\nonto a file that is already two-thirds written.\r\n\r\n- **`extra_stall_s` (default 180)** stops a worker that has produced nothing for\r\n that long. It is reset by observable progress — the worker's stream, its\r\n stderr, or a declared file that changed on disk — so it never fires on one that\r\n is merely slow. `stalled` is retryable, which is what makes the resume ORC's\r\n own spelling of typing `continue`.\r\n- **`extra_fallback_agent` (default `band`)** decides who picks the task up.\r\n `ask` stops and puts the menu to you; any installed agent name pins one. It\r\n changes *who*, never the score, the declared files or the acceptance criteria.\r\n- **Every dispatch writes its own spend record**, so a cost report never depends\r\n on a run remembering to narrate what it spent. `orc extra stats` merges the\r\n spend log, the traces and saved returns, and always says how many rows came\r\n from each.\r\n\r\n**`/orc-quick` is inert here** and announces it — that lane asks which agent\r\nbefore every dispatch, so no setting may pre-answer it. **`/orc-challenge` never\r\nroutes foreign**: swapping a lens for a different model does not make the lane\r\ncheaper, it changes what is being measured.\r\n\r\n**The whole subsystem, with every command and key:\r\n[guides/extra-models.md](guides/extra-models.md).**\r\n\r\n---\r\n\r\n## How the model is picked\r\n\r\nEach task is scored 0–100 by **arithmetic, not judgement**: the planner reports\r\nfacets (breadth, novelty, logic, test surface, cited risk, uncertainty) and a\r\nfixed published formula turns them into a number. A cited risk forces a floor of\r\n70. The score maps through a published table to a **named, model-pinned agent**,\r\nso what ran is inspectable rather than requested in prose.\r\n\r\n> **The rule that catches everyone:** a subagent's model can never be higher than\r\n> your main session's. Run your session on Opus 5.\r\n\r\n**Full detail — the bands, `opus5_only`, and the tier guard `orc init` installs:\r\n[guides/model-selection.md](guides/model-selection.md).**\r\n\r\n---\r\n\r\n## Configuration\r\n\r\nSettings are edited with the **`orc config` CLI** — deterministic terminal I/O,\r\nso it costs **zero model tokens**.\r\n\r\n```bash\r\norc config # interactive menu\r\norc config list # the effective config\r\norc config recommend # read this repo, suggest ONE profile, with reasons\r\norc config profile paranoid\r\n```\r\n\r\nYour changes live in `.claude/orc.config.yaml`, which `orc update` never\r\nclobbers. `orc ui` ▸ Settings edits the same keys through the same validators.\r\n\r\n**Every key, with defaults and what each one does:\r\n[guides/configuration.md](guides/configuration.md).**\r\n\r\n---\r\n\r\n## What is inside the package\r\n\r\n```\r\ntemplates/\r\n├── skills/ 31 skill folders, 38 SKILL.md files (a lane may ship subskills) —\r\n│ the lanes above, plus the ones with no command of their own:\r\n│ context-combiner, orc-advisor, orc-judge, orc-analyze-mini,\r\n│ and _shared/ (cross-lane contract prose)\r\n├── commands/ 29 slash commands\r\n├── hooks/ effort guard (PreToolUse) · statusline warning · behavior trace\r\n└── agents/ 48 model-pinned subagents + MODEL-MAPPING.md\r\nbin/cli.js installer, config editor, flow composer, run-state reader, and\r\n the deterministic half of every lane. Every read speaks --json\r\nbin/graph*.js the code graph: store, extraction, resolution + its cache, signals, notes\r\nbin/webui/ `orc ui` — the local control panel: css/ + js/ + i18n// +\r\n fixtures/, one file per layer and per panel. Zero deps, no build step\r\nbin/mockrun-catalog.js the mocked-run catalogue (derived from the files on disk)\r\nmock-run/ the mocked runs themselves — start at INDEX.md\r\nguides/ configuration · model selection · documents · knowledge reads · other AI models\r\n```\r\n\r\nThe `orc` skill is a thin **spine**: it loads a reference or a subskill only when\r\nthat phase actually runs, so a small task never pays for the machinery of a big\r\none.\r\n\r\n---\r\n\r\n## Longer guides\r\n\r\nSome lanes ship a full how-to next to the skill, in plain language:\r\n\r\n| Guide | Read it when |\r\n|---|---|\r\n| [ORC-QUICK](templates/skills/orc-quick/README.md) | you want the quick lane's complete worked runs |\r\n| [ORC-DIY](templates/skills/orc-diy/README.md) | you want to compose your own lane |\r\n| [ORC-WIKI](templates/skills/orc-wiki/README.md) | you want the knowledge base, and cross-repo crosslink setup |\r\n| [ORC-PR-SETUP](templates/skills/orc-pr-setup/README.md) | you want to split a big change into stacked PRs |\r\n| [ORC-PR-DRIVER](templates/skills/orc-pr-driver/README.md) | you have a stack plan and want to build, submit and merge it |\r\n| [Rules](guides/rules.md) | you want the 65 anti-slop rules, the precedence ladder and the lint |\r\n| [Configuration](guides/configuration.md) · [Model selection](guides/model-selection.md) | you want every key, or the scoring bands |\r\n| [Other AI models](guides/extra-models.md) | you want part of the ladder to run somewhere other than Claude |\r\n\r\nEvery skill also ships its own `SKILL.md` and `references/`. The guides above are\r\nthe human-facing versions.\r\n\r\n---\r\n\r\n## Eval status\r\n\r\nThe constellation is graded **end to end**, not file by file: one executable\r\nspec per lane, run against a sandboxed Express fixture, graded from on-disk\r\nevidence — behavior traces, run folders and artifacts.\r\n\r\nThe last full round was the **30-eval suite against the v0.34.0 payload**: 25\r\nfilled result files and 38 trace files, with 5 evals never graded and 2 only\r\npartly graded — all named in the report. Everything found there was either fixed\r\nin a later release or is still listed. Read it as a record of that round, not as\r\na current audit: [EVAL-REPORT.md](EVAL-REPORT.md).\r\n\r\n---\r\n\r\n## Design principles\r\n\r\n- **Never implement at the top.** The orchestrator coordinates; scored subagents\r\n do the work.\r\n- **Bound the scope before parallelizing.** A misunderstanding is cheap to fix\r\n before five agents build on it.\r\n- **Disk over memory.** Every pause is a clean resume point.\r\n- **Pinned, inspectable models.** Named agents, models in frontmatter.\r\n- **Your codebase wins.** Learned patterns defer to your project; only security\r\n and correctness invariants are non-negotiable.\r\n- **Additive knowledge.** The wiki improves planning when it is there and costs\r\n nothing when it is not.\r\n- **Say what you do not know.** `UNCHECKABLE`, `no card`, `insufficient history`\r\n are real answers. A confident guess is worse than an honest gap.\r\n\r\n---\r\n\r\n## Changelog\r\n\r\n**Full history: [CHANGELOG.md](CHANGELOG.md)** — or `orc changelog`, which prints\r\nonly what is newer than the version you have.\r\n\r\n### v1.8.1 - the guard that only failed on Windows _(2026-09-16)_\r\n\r\n`npm publish` could not run on Windows. The `prepack` guard reported nine\r\ndrifted executor agents, and `git status` reported no change at all.\r\n\r\n**The cause was a line ending, not content.** `.gitattributes` says\r\n`* text=auto`, so git stores the generated executor agents with LF and checks\r\nthem out native — a Windows clone holds CRLF. The generator read the template\r\nas it is on disk, but injected the `effort:` frontmatter line with a hardcoded\r\nLF. `bin/build-agents.js` now takes its line ending from the template, and\r\n`--check` compares LF-normalized text, so the guard gives the same result in\r\neither worktree. A test in `test/payload.test.js` holds the rule.\r\n\r\nOne test carried the same fault. `npm test` also runs on `prepack`, and the\r\nstatusline graph-component test sliced a hook's source to the next `\\n}\\n` — a\r\nneedle a CRLF checkout never contains. It now normalizes the source first.\r\n\r\nNo shipped payload changed. The generated agent files are byte-identical.\r\n\r\n
\r\nEarlier releases — 117 of them, titles only. Full text in CHANGELOG.md.\r\n\r\n- **v1.8.0** — the code graph: a map of the code that stays fresh · _2026-09-16_\r\n- **v1.7.1** — the rules card now reaches the agent · _2026-09-14_\r\n- **v1.7.0** — the rules that keep the slop out · _2026-09-13_\r\n- **v1.6.0** — the rule that can finally say no · _2026-09-07_\r\n- **v1.5.0** — the lane that runs the test · _2026-09-07_\r\n- **v1.4.2** — the panel that stops reloading, and the fields you could not read back · _2026-09-05_\r\n- **v1.4.1** — the board you can actually use · _2026-09-05_\r\n- **v1.4.0** — the agent panel, and the number that was missing · _2026-09-04_\r\n- **v1.3.0** — build your own status line · _2026-09-04_\r\n- **v1.2.1** — the status line says what ORC is doing · _2026-09-04_\r\n- **v1.2.0** — a retry that cloned the agent, and a window you can watch empty · _2026-09-04_\r\n- **v1.1.0** — the wait, and a window ORC can finally see · _2026-08-31_\r\n- **v1.0.0** — config, phases and calls stop being prose · _2026-08-30_\r\n- **v0.56.1** — a worker that is alive and doing nothing · _2026-08-28_\r\n- **v0.56.0** — a rename moved the command, and nobody could reach the fix · _2026-08-27_\r\n- **v0.55.2** — a gate that is never probed is a gate that is always off · _2026-08-27_\r\n- **v0.55.1** — ORC is on npm · _2026-08-27_\r\n- **v0.55.0** — a score is what a band needs, and four lanes do not have one · _2026-08-26_\r\n- **v0.54.0** — a failed dispatch is a POSITION, not a blank page · _2026-08-25_\r\n- **v0.53.4** — the reload that dropped its own token · _2026-08-24_\r\n- **v0.53.3** — the key it never sent · _2026-08-24_\r\n- **v0.53.2** — the cost that was paid and never written down · _2026-08-24_\r\n- **v0.53.1** — \"up to date\" now names what it checked · _2026-08-23_\r\n- **v0.53.0** — the schema the provider rejected, and a routing table you can read · _2026-08-23_\r\n- **v0.52.0** — the connection that could not be used, and the routing nobody could see · _2026-08-23_\r\n- **v0.51.0** — the tools you already have, and a connection that proves itself · _2026-08-22_\r\n- **v0.50.0** — work that runs somewhere else · _2026-08-22_\r\n- **v0.49.5** — house rules are text, and the hand-back writes itself · _2026-08-21_\r\n- **v0.49.4** — the panel was being handed half an answer · _2026-08-20_\r\n- **v0.49.3** — coverage on a large repo · _2026-08-19_\r\n- **v0.49.2** — house rules, a run map before you pay, and three defects · _2026-08-18_\r\n- **v0.49.1** — the challenge council, and a `--json` that stops throwing things away · _2026-08-18_\r\n- **v0.49.0** — the document is a folder, and the file is a build artifact · _2026-08-17_\r\n- **v0.48.1** — one file per thing, and a document that can be finished · _2026-08-16_\r\n- **v0.48.0** — a document long enough to end a session, written anyway · _2026-08-13_\r\n- **v0.47.0** — the lane that refuses to produce · _2026-08-12_\r\n- **v0.46.1** — see a lane run before you pay for one · _2026-08-12_\r\n- **v0.46.0** — a lane that remembers, a lane that declines, and a lane that measures · _2026-08-10_\r\n- **v0.45.0** — `/orc-brainstorm`: for when you do not have the idea yet · _2026-08-10_\r\n- **v0.44.1** — apply when you say so, and a spotlight that survives a banner · _2026-08-09_\r\n- **v0.44.0** — the panel stops making you type what it already knows · _2026-08-09_\r\n- **v0.43.7** — the flow you can see, and a boundary you can read · _2026-08-09_\r\n- **v0.43.6** — `orc ui` in two languages, and panels that point at the right page · _2026-08-08_\r\n- **v0.43.5** — the update check works, and the UI teaches itself · _2026-08-08_\r\n- **v0.43.4** — a warning that finally clears, an Experiment panel, crosslink from the UI · _2026-08-08_\r\n- **v0.43.3** — `orc ui`: it tells you about updates, and 36 keys stop being a wall · _2026-08-08_\r\n- **v0.43.2** — `orc ui`: boxes stop colliding, because the container owns the gap · _2026-08-08_\r\n- **v0.43.1** — the panel's stylesheet and script actually reach the browser · _2026-08-08_\r\n- **v0.43.0** — `orc ui`: a control panel for everything that is not ai · _2026-08-08_\r\n- **v0.42.0** — Say what you mean, see what it costs, find your way back · _2026-08-08_\r\n- **v0.41.0** — A wiki that can tell you it is fresh, and TDD only where it can fail · _2026-08-06_\r\n- **v0.40.0** — Gotchas: repair memory that outlives the run · _2026-08-06_\r\n- **v0.39.0** — The read ladder, and foreign input that is evidence rather than instruction · _2026-08-06_\r\n- **v0.38.1** — `orc doctor --json` + handoff carry-over that says what is re-derived · _2026-08-06_\r\n- **v0.38.0** — `/orc-quick`: the quick lane, and the gate no config can collapse · _2026-08-05_\r\n- **v0.37.0** — Stacked pull requests: a measured ship gate + two standalone lanes · _2026-08-03_\r\n- **v0.36.0** — `opus5_only`: one model for every role, not just executors · _2026-08-02_\r\n- **v0.35.0** — `opus5_executor_only`: one model, effort as the cost dial · _2026-08-02_\r\n- **v0.34.8** — `orc pattern status` rejects a language key the payload has never heard of · _2026-08-01_\r\n- **v0.34.7** — DIY: a usable status contract, and compile docs that match the compiler · _2026-08-01_\r\n- **v0.34.6** — Analyze: the evidence gate now covers the rows a good analysis produces · _2026-08-01_\r\n- **v0.34.5** — Wiki: stop losing tags silently, let a delta clear its own delta · _2026-08-01_\r\n- **v0.34.4** — Planner: scorable facets, and TDD rules scoped to reality · _2026-08-01_\r\n- **v0.34.3** — Slice boundary: the worktree, not the editor · _2026-08-01_\r\n- **v0.34.2** — Trace subsystem: the pointer clobber, and a writer contract that holds · _2026-08-01_\r\n- **v0.34.1** — Install integrity: run state survives `orc update` · _2026-08-01_\r\n- **v0.34.0** — Opus 5: top scoring band, every core role, medium-effort session tier · _2026-07-25_\r\n- **v0.33.0** — Knowledge deepening + verification revamp · _2026-07-25_\r\n- **v0.32.0** — Trace revamp: narration is dispatched, not remembered · _2026-07-24_\r\n- **v0.31.0** — Execution-integrity revamp: plan handoff, attributable traces, facet scoring · _2026-07-23_\r\n- **v0.30.0** — Scoring revamp, Fable 5 role override, tier-aware guards, `orc onboarding` · _2026-07-23_\r\n- **v0.29.0** — Drift-prevention hardening: install manifest + prune, `orc doctor`, a real test suite · _2026-07-22_\r\n- **v0.28.1** — Defect fixes: package encoding, trace event routing, count/doc drift · _2026-07-22_\r\n- **v0.28.0** — Run integrity: rich full-lane traces, deterministic wave stop, visible knowledge gates · _2026-07-21_\r\n- **v0.27.0** — `/orc-poly`: plan one change across two-or-more repos without drift · _2026-07-20_\r\n- **v0.26.0** — Test-gen output pinned to a visible `test-generator//` deliverable · _2026-07-19_\r\n- **v0.25.1** — Eval report: the full 17-lane suite graded against the v0.25.0 payload · _2026-07-18_\r\n- **v0.25.0** — Deterministic artifact detection: a generated wiki/pattern is never missed · _2026-07-18_\r\n- **v0.24.0** — Crosslink fused into wiki generation: always-on, per-scan-task, never wiped · _2026-07-18_\r\n- **v0.23.0** — Trace fix: SPAWN restored on the `Agent` tool, stale runs rotate to fresh files · _2026-07-18_\r\n- **v0.22.0** — `/orc-learn`: per-feature onboarding docs — learning.md + knowledge.md, wiki-deep, git-ignored · _2026-07-17_\r\n- **v0.21.0** — Statusline shows live subscription usage: 5h ↔ weekly, official numbers · _2026-07-16_\r\n- **v0.20.0** — One source of truth: generated executor agents + shared cross-lane contracts · _2026-07-16_\r\n- **v0.19.0** — Thin spines: skill compaction, budget lint, and a trace that logs every phase · _2026-07-16_\r\n- **v0.18.0** — `orc wiki sync`: the wiki registers itself — a paused scan is no longer an invisible wiki · _2026-07-15_\r\n- **v0.17.3** — Trace the wiki consult: Phase 1 now logs whether the run grounded in the wiki (and if it was stale) · _2026-07-14_\r\n- **v0.17.2** — Behavior-trace logging is permanent + the trace folder is now created deterministically · _2026-07-14_\r\n- **v0.17.1** — Complete cross-repo crosslink setup guide in the orc-wiki README · _2026-07-14_\r\n- **v0.17.0** — `orc crosslink`: cross-repo wiki references — advisory boundary contracts · _2026-07-14_\r\n- **v0.16.1** — Interactive `orc diy` composer + numbered picks in `orc config` · _2026-07-14_\r\n- **v0.16.0** — `/orc-diy`: build your own lane — CLI-composed flow, compiled, hard-gated · _2026-07-14_\r\n- **v0.15.0** — Wiki v2: evidence-anchored docs · per-file staleness registry · integrity gate · _2026-07-14_\r\n- **v0.14.0** — Postgres data-access playbook: cross-cutting query grounding · _2026-07-13_\r\n- **v0.13.0** — `/orc-claude`: local CLAUDE.md builder — fenced sections, fingerprint refresh, zero questions · _2026-07-12_\r\n- **v0.12.0** — Lossless context-combiner: conservation gate · overlap taxonomy · evidence freshness · _2026-07-12_\r\n- **v0.11.0** — `/orc-fast`: knowledge-gated speed lane + wiki freshness infrastructure · _2026-07-12_\r\n- **v0.10.1** — README: a fuller \"Why ORC exists\" · _2026-07-12_\r\n- **v0.10.0** — `/orc-ultra`: max-effort advisor + three judgment gates for ultra-complex work · _2026-07-12_\r\n- **v0.9.0** — Trust-but-verify the analyst→planner chain: quote-anchored evidence · coverage gate · anchored judgment · _2026-07-12_\r\n- **v0.8.1** — /orc-retro delivers upstream: PR/issue to the ORC repo, channel-gated · _2026-07-12_\r\n- **v0.8.0** — Close the loop: grounded intake · scoring anchors · OUTCOME marker · /orc-retro trace miner · eval harness · _2026-07-12_\r\n- **v0.7.0** — Evidence everywhere: grounded plans · verbatim proof · anchored findings · contract lint · trace fixes · _2026-07-12_\r\n- **v0.6.0** — P0–P3 ladder · house rules · deep playbooks + wired gates · 3 new languages · FE rule packs · security pass · _2026-07-11_\r\n- **v0.5.1** — Statusline false-degrade fix · _2026-07-11_\r\n- v0.5.0 — Code-pattern findings: executors match your house style, invariants always enforced\r\n- v0.4.5 — Rewrite weak worker descriptions (the real score lever)\r\n- v0.4.4 — Act on external review: raise sub-70 workers, fix cross-spine paths\r\n- v0.4.3 — `orc-analyze`: trim description under the 1024-char skill-spec limit\r\n- v0.4.2 — External-review pass: worked examples + sharper mini-analyst activation\r\n- v0.4.1 — `orc-mini`: faster, safer fast-lane — smoke gate, opt-in tests, trimmed ceremony\r\n- v0.4.0 — Opt-in Phase 6.5 Test Authoring (writes test cases, never runs them)\r\n- v0.3.0 — Opt-in behavior-trace logging + claimed-vs-actual model verification\r\n- v0.2.4 — `orc-analyze`: gather anchored adjacent-scope context (non-actionable)\r\n- v0.2.3 — Context Combiner: merge 2+ related analyses into one combined spec\r\n- v0.2.2 — Config: enforce per-key override-first resolution\r\n- v0.2.1 — Move config editing into the `orc config` CLI (zero-token); drop `/orc-config`\r\n- v0.2.0 — Doc-optional evidence-backed analyst + deep mode\r\n\r\n
\r\n\r\n## Requirements\r\n\r\n- **Claude Code** — it reads the skills, commands and agents.\r\n- **Node 18+** — for the installer only. The skills themselves have zero\r\n dependencies.\r\n\r\n## License\r\n\r\nMIT — the `license` field in `package.json` is the canonical statement.\r\n","readmeFilename":"README.md"}