{"_id":"@calculator53295/knowledge-arena","name":"@calculator53295/knowledge-arena","dist-tags":{"latest":"0.1.0"},"versions":{"0.1.0":{"name":"@calculator53295/knowledge-arena","version":"0.1.0","description":"Deterministic local-model and retrieval-strategy benchmark runner","type":"module","sideEffects":false,"exports":{".":{"types":"./src/index.d.ts","default":"./src/index.js"}},"bin":{"knowledge-arena":"bin/knowledge-arena.js"},"types":"./src/index.d.ts","scripts":{"test":"node --test test/*.test.js"},"engines":{"node":">=20"},"publishConfig":{"access":"public"},"repository":{"type":"git","url":"git+https://github.com/Calculator5329/agent-colosseum.git","directory":"packages/knowledge-arena"},"homepage":"https://github.com/Calculator5329/agent-colosseum/tree/main/packages/knowledge-arena#readme","bugs":{"url":"https://github.com/Calculator5329/agent-colosseum/issues"},"license":"MIT","_id":"@calculator53295/knowledge-arena@0.1.0","_integrity":"sha512-9usHxd01js0y+g40+BgzozDqBv6HyybPk3mVwmyeusOkPEzObV/pfAOC0S2xqSpA/cyepJcH1BD3heakg0op6w==","_resolved":"/home/ethan/projects/ai/agent-colosseum/_release/npm/calculator53295-knowledge-arena-0.1.0.tgz","_from":"file:/home/ethan/projects/ai/agent-colosseum/_release/npm/calculator53295-knowledge-arena-0.1.0.tgz","_nodeVersion":"24.18.0","_npmVersion":"11.16.0","dist":{"integrity":"sha512-9usHxd01js0y+g40+BgzozDqBv6HyybPk3mVwmyeusOkPEzObV/pfAOC0S2xqSpA/cyepJcH1BD3heakg0op6w==","shasum":"24e11928624e8f2d1529259d4f9f759b354fcad0","tarball":"https://registry.npmjs.org/@calculator53295/knowledge-arena/-/knowledge-arena-0.1.0.tgz","fileCount":9,"unpackedSize":21086,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCICR5kkBHQ+/oHhZHzzOW2hwlx7dMikFuIMS6gIQ72PkdAiEAqyjRp27nM4YnMm818081kdnDPLHYhIXRlFt0AmnWozg="}]},"_npmUser":{"name":"calculator53295","email":"5329548871.eg@gmail.com"},"directories":{},"maintainers":[{"name":"calculator53295","email":"5329548871.eg@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/knowledge-arena_0.1.0_1783897241642_0.8530067585432011"},"_hasShrinkwrap":false}},"time":{"created":"2026-07-12T23:00:41.501Z","0.1.0":"2026-07-12T23:00:41.776Z","modified":"2026-07-12T23:00:41.979Z"},"maintainers":[{"name":"calculator53295","email":"5329548871.eg@gmail.com"}],"description":"Deterministic local-model and retrieval-strategy benchmark runner","homepage":"https://github.com/Calculator5329/agent-colosseum/tree/main/packages/knowledge-arena#readme","repository":{"type":"git","url":"git+https://github.com/Calculator5329/agent-colosseum.git","directory":"packages/knowledge-arena"},"bugs":{"url":"https://github.com/Calculator5329/agent-colosseum/issues"},"license":"MIT","readme":"# @calculator53295/knowledge-arena\n\nZero-dependency benchmark runner for local models across offline retrieval\nstrategies and approved public database schemas. It consumes the local-ai-lab\nplugin API and local Ollama only; no remote provider is configured.\n\n```bash\nknowledge-arena plan benchmarks/knowledge/matrix.json benchmarks/knowledge/tasks.json\nknowledge-arena probe benchmarks/knowledge/tasks.json \\\n  benchmarks/knowledge/results/corpus-v1.probes.json --execute\nknowledge-arena run benchmarks/knowledge/matrix.json benchmarks/knowledge/tasks.json \\\n  benchmarks/knowledge/results/round-001.trials.json --execute\nknowledge-arena score benchmarks/knowledge/results/round-001.trials.json \\\n  benchmarks/knowledge/results/round-001.scorecard.json\n```\n\n`run` requires an explicit `--execute` and refuses to overwrite evidence.\nScoring remains disaggregated: source hit (40), expected-term recall (35), and\ncitation validity (25), plus separately measured retrieval, generation, and\nend-to-end latency. Database trials use only public schema endpoints. Private\nand restricted aliases are absent from the corpus and cannot be selected.\nRetrieved passage bodies are transient and never written to evidence. Stored\nURIs remove the user home prefix, while answer text redacts email addresses and\ndollar-denominated figures before persistence.\n\nMatrices may set `repetitions` (1–10) and an integer `seed`. Each repetition\nuses `temperature: 0` and a stable incremented seed. Scorecards include 95%\nnormal-approximation intervals for model, strategy, dataset, and exact-cell\ngroups; small samples must be interpreted cautiously.\n","readmeFilename":"README.md","_rev":"1-3458b135d00a1b02cdbe957a1b400236"}