{"id":"@lssm-tech/lib.agent-evals","slug":"lib.agent-evals","name":"@lssm-tech/lib.agent-evals","version":"2.2.4","path":"packages/platform/agents/agent-evals","manifestPath":"packages/platform/agents/agent-evals/package.json","manifestBlobSha":"41ce3523298eef09aebe7e5e1c27b625067ad688","layer":"platform","physicalArea":"agents","primaryArea":"observability-and-testing","classification":{"privateDeclared":null,"catalogFamily":"sharedLibs","catalogKind":"library","catalogStatus":"active","adoptionMode":"adoptable","stability":"experimental"},"description":"Agent evaluation runners, adapters, graders, and recommendation evidence.","descriptionOrigin":"package.json","catalog":{"family":"sharedLibs","kind":"library","status":"active","adoptionMode":"adoptable","stability":"experimental","platforms":["server"],"runtimes":["node"],"entrypoints":[".","./adapters","./adapters/live-model","./adapters/llm-judge","./adapters/scripted-model","./agent-evals.feature","./digest","./graders","./graders/artifact","./graders/environment-outcome","./graders/exact-match","./graders/policy","./graders/schema","./metrics-v2","./model-canary","./model-canary-v2","./model-canary-verification-v2","./model-canary-verification","./model-qualification","./model-qualification-integrity","./model-qualification-integrity-v2","./model-qualification-lifecycle-v2","./model-qualification-v2","./runner","./runner-v2","./types","./types-v2"],"documentation":[{"title":"Lib Agent Evals documentation","url":"https://www.npmjs.com/package/%40lssm-tech%2Flib.agent-evals","kind":"readme"}],"examples":[],"replacements":[],"capabilities":[{"id":"package.lssm.tech.lib.agent.evals","title":"Lib Agent Evals","description":"Agent evaluation runners, adapters, graders, and recommendation evidence.","status":"active","useCases":["Adopt agent evaluation runners, adapters, graders, and recommendation evidence"],"aliases":["contractspec","agent","evals","grader","evaluation","typescript","lib"],"pythonEquivalents":[],"canonicalImports":[{"from":"@lssm-tech/lib.agent-evals","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/adapters","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/adapters/live-model","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/adapters/llm-judge","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/adapters/scripted-model","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/agent-evals.feature","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/graders","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/graders/artifact","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/graders/environment-outcome","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/graders/exact-match","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/graders/policy","symbols":[]},{"from":"@lssm-tech/lib.agent-evals/graders/schema","symbols":[]}],"constraints":{"platforms":["server"],"runtimes":["node"],"avoidWhen":["The @lssm-tech/lib.agent-evals package constraints do not match the target runtime or platform."],"requires":["TypeScript"]},"alternatives":[],"relationships":[],"resolutionPriority":50}]},"canonicalImports":[{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/adapters","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/adapters/live-model","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/adapters/llm-judge","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/adapters/scripted-model","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/agent-evals.feature","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/graders","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/graders/artifact","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/graders/environment-outcome","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/graders/exact-match","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/graders/policy","symbols":[]},{"capabilityId":"package.lssm.tech.lib.agent.evals","from":"@lssm-tech/lib.agent-evals/graders/schema","symbols":[]}],"scripts":{"publish:pkg":"bun publish --tolerate-republish --ignore-scripts --verbose","publish:pkg:canary":"bun publish:pkg --tag canary","build":"contractspec-bun-build build","build:bundle":"contractspec-bun-build transpile","build:types":"contractspec-bun-build types","dev":"contractspec-bun-build dev","clean":"rimraf dist .turbo","lint":"bun run lint:fix","lint:fix":"node ../../../../scripts/biome.cjs check --write --unsafe --only=nursery/useSortedClasses . && node ../../../../scripts/biome.cjs check --write .","lint:check":"node ../../../../scripts/biome.cjs check .","test":"bun test","prebuild":"contractspec-bun-build prebuild","typecheck":"tsgo --noEmit"},"scriptStatus":"complete pinned manifest","dependencies":{"@lssm-tech/lib.agent-protocol-spec":"workspace:^","@lssm-tech/lib.assurance-spec":"workspace:^","@lssm-tech/lib.contracts-spec-core":"workspace:^"},"devDependencies":{"@lssm-tech/tool.bun":"workspace:*","@lssm-tech/tool.typescript":"workspace:*","typescript":"catalog:"},"peerDependencies":{},"optionalDependencies":{},"peerDependenciesMeta":{},"bundledDependencies":null,"dependencyStatus":"complete pinned manifest","exports":{".":"./src/index.ts","./adapters":"./src/adapters/index.ts","./adapters/live-model":"./src/adapters/live-model.ts","./adapters/llm-judge":"./src/adapters/llm-judge.ts","./adapters/scripted-model":"./src/adapters/scripted-model.ts","./agent-evals.feature":"./src/agent-evals.feature.ts","./digest":"./src/digest.ts","./graders":"./src/graders/index.ts","./graders/artifact":"./src/graders/artifact.ts","./graders/environment-outcome":"./src/graders/environment-outcome.ts","./graders/exact-match":"./src/graders/exact-match.ts","./graders/policy":"./src/graders/policy.ts","./graders/schema":"./src/graders/schema.ts","./metrics-v2":"./src/metrics-v2.ts","./model-canary":"./src/model-canary.ts","./model-canary-v2":"./src/model-canary-v2.ts","./model-canary-verification":"./src/model-canary-verification/index.ts","./model-canary-verification-v2":"./src/model-canary-verification-v2.ts","./model-qualification":"./src/model-qualification.ts","./model-qualification-integrity":"./src/model-qualification-integrity.ts","./model-qualification-integrity-v2":"./src/model-qualification-integrity-v2.ts","./model-qualification-lifecycle-v2":"./src/model-qualification-lifecycle-v2.ts","./model-qualification-v2":"./src/model-qualification-v2.ts","./runner":"./src/runner.ts","./runner-v2":"./src/runner-v2.ts","./types":"./src/types.ts","./types-v2":"./src/types-v2.ts"},"publishExports":{".":{"types":"./dist/index.d.ts","browser":"./dist/browser/index.js","bun":"./dist/index.js","node":"./dist/node/index.js","default":"./dist/index.js"},"./adapters":{"types":"./dist/adapters/index.d.ts","browser":"./dist/browser/adapters/index.js","bun":"./dist/adapters/index.js","node":"./dist/node/adapters/index.js","default":"./dist/adapters/index.js"},"./adapters/live-model":{"types":"./dist/adapters/live-model.d.ts","browser":"./dist/browser/adapters/live-model.js","bun":"./dist/adapters/live-model.js","node":"./dist/node/adapters/live-model.js","default":"./dist/adapters/live-model.js"},"./adapters/llm-judge":{"types":"./dist/adapters/llm-judge.d.ts","browser":"./dist/browser/adapters/llm-judge.js","bun":"./dist/adapters/llm-judge.js","node":"./dist/node/adapters/llm-judge.js","default":"./dist/adapters/llm-judge.js"},"./adapters/scripted-model":{"types":"./dist/adapters/scripted-model.d.ts","browser":"./dist/browser/adapters/scripted-model.js","bun":"./dist/adapters/scripted-model.js","node":"./dist/node/adapters/scripted-model.js","default":"./dist/adapters/scripted-model.js"},"./agent-evals.feature":{"types":"./dist/agent-evals.feature.d.ts","browser":"./dist/browser/agent-evals.feature.js","bun":"./dist/agent-evals.feature.js","node":"./dist/node/agent-evals.feature.js","default":"./dist/agent-evals.feature.js"},"./digest":{"types":"./dist/digest.d.ts","browser":"./dist/browser/digest.js","bun":"./dist/digest.js","node":"./dist/node/digest.js","default":"./dist/digest.js"},"./graders":{"types":"./dist/graders/index.d.ts","browser":"./dist/browser/graders/index.js","bun":"./dist/graders/index.js","node":"./dist/node/graders/index.js","default":"./dist/graders/index.js"},"./graders/artifact":{"types":"./dist/graders/artifact.d.ts","browser":"./dist/browser/graders/artifact.js","bun":"./dist/graders/artifact.js","node":"./dist/node/graders/artifact.js","default":"./dist/graders/artifact.js"},"./graders/environment-outcome":{"types":"./dist/graders/environment-outcome.d.ts","browser":"./dist/browser/graders/environment-outcome.js","bun":"./dist/graders/environment-outcome.js","node":"./dist/node/graders/environment-outcome.js","default":"./dist/graders/environment-outcome.js"},"./graders/exact-match":{"types":"./dist/graders/exact-match.d.ts","browser":"./dist/browser/graders/exact-match.js","bun":"./dist/graders/exact-match.js","node":"./dist/node/graders/exact-match.js","default":"./dist/graders/exact-match.js"},"./graders/policy":{"types":"./dist/graders/policy.d.ts","browser":"./dist/browser/graders/policy.js","bun":"./dist/graders/policy.js","node":"./dist/node/graders/policy.js","default":"./dist/graders/policy.js"},"./graders/schema":{"types":"./dist/graders/schema.d.ts","browser":"./dist/browser/graders/schema.js","bun":"./dist/graders/schema.js","node":"./dist/node/graders/schema.js","default":"./dist/graders/schema.js"},"./metrics-v2":{"types":"./dist/metrics-v2.d.ts","browser":"./dist/browser/metrics-v2.js","bun":"./dist/metrics-v2.js","node":"./dist/node/metrics-v2.js","default":"./dist/metrics-v2.js"},"./model-canary":{"types":"./dist/model-canary.d.ts","browser":"./dist/browser/model-canary.js","bun":"./dist/model-canary.js","node":"./dist/node/model-canary.js","default":"./dist/model-canary.js"},"./model-canary-v2":{"types":"./dist/model-canary-v2.d.ts","browser":"./dist/browser/model-canary-v2.js","bun":"./dist/model-canary-v2.js","node":"./dist/node/model-canary-v2.js","default":"./dist/model-canary-v2.js"},"./model-canary-verification":{"types":"./dist/model-canary-verification/index.d.ts","browser":"./dist/browser/model-canary-verification/index.js","bun":"./dist/model-canary-verification/index.js","node":"./dist/node/model-canary-verification/index.js","default":"./dist/model-canary-verification/index.js"},"./model-canary-verification-v2":{"types":"./dist/model-canary-verification-v2.d.ts","browser":"./dist/browser/model-canary-verification-v2.js","bun":"./dist/model-canary-verification-v2.js","node":"./dist/node/model-canary-verification-v2.js","default":"./dist/model-canary-verification-v2.js"},"./model-qualification":{"types":"./dist/model-qualification.d.ts","browser":"./dist/browser/model-qualification.js","bun":"./dist/model-qualification.js","node":"./dist/node/model-qualification.js","default":"./dist/model-qualification.js"},"./model-qualification-integrity":{"types":"./dist/model-qualification-integrity.d.ts","browser":"./dist/browser/model-qualification-integrity.js","bun":"./dist/model-qualification-integrity.js","node":"./dist/node/model-qualification-integrity.js","default":"./dist/model-qualification-integrity.js"},"./model-qualification-integrity-v2":{"types":"./dist/model-qualification-integrity-v2.d.ts","browser":"./dist/browser/model-qualification-integrity-v2.js","bun":"./dist/model-qualification-integrity-v2.js","node":"./dist/node/model-qualification-integrity-v2.js","default":"./dist/model-qualification-integrity-v2.js"},"./model-qualification-lifecycle-v2":{"types":"./dist/model-qualification-lifecycle-v2.d.ts","browser":"./dist/browser/model-qualification-lifecycle-v2.js","bun":"./dist/model-qualification-lifecycle-v2.js","node":"./dist/node/model-qualification-lifecycle-v2.js","default":"./dist/model-qualification-lifecycle-v2.js"},"./model-qualification-v2":{"types":"./dist/model-qualification-v2.d.ts","browser":"./dist/browser/model-qualification-v2.js","bun":"./dist/model-qualification-v2.js","node":"./dist/node/model-qualification-v2.js","default":"./dist/model-qualification-v2.js"},"./runner":{"types":"./dist/runner.d.ts","browser":"./dist/browser/runner.js","bun":"./dist/runner.js","node":"./dist/node/runner.js","default":"./dist/runner.js"},"./runner-v2":{"types":"./dist/runner-v2.d.ts","browser":"./dist/browser/runner-v2.js","bun":"./dist/runner-v2.js","node":"./dist/node/runner-v2.js","default":"./dist/runner-v2.js"},"./types":{"types":"./dist/types.d.ts","browser":"./dist/browser/types.js","bun":"./dist/types.js","node":"./dist/node/types.js","default":"./dist/types.js"},"./types-v2":{"types":"./dist/types-v2.d.ts","browser":"./dist/browser/types-v2.js","bun":"./dist/types-v2.js","node":"./dist/node/types-v2.js","default":"./dist/types-v2.js"}},"types":"./dist/index.d.ts","files":["dist","README.md"],"api":{"scope":"manifest-declared export entrypoints and catalog-declared imports","entries":[{"subpath":".","importPath":"@lssm-tech/lib.agent-evals","target":"./src/index.ts","targets":[{"conditions":[],"target":"./src/index.ts","sourcePath":"packages/platform/agents/agent-evals/src/index.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/index.ts","sourceStatus":"verified source file"}]},{"subpath":"./adapters","importPath":"@lssm-tech/lib.agent-evals/adapters","target":"./src/adapters/index.ts","targets":[{"conditions":[],"target":"./src/adapters/index.ts","sourcePath":"packages/platform/agents/agent-evals/src/adapters/index.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/adapters/index.ts","sourceStatus":"verified source file"}]},{"subpath":"./adapters/live-model","importPath":"@lssm-tech/lib.agent-evals/adapters/live-model","target":"./src/adapters/live-model.ts","targets":[{"conditions":[],"target":"./src/adapters/live-model.ts","sourcePath":"packages/platform/agents/agent-evals/src/adapters/live-model.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/adapters/live-model.ts","sourceStatus":"verified source file"}]},{"subpath":"./adapters/llm-judge","importPath":"@lssm-tech/lib.agent-evals/adapters/llm-judge","target":"./src/adapters/llm-judge.ts","targets":[{"conditions":[],"target":"./src/adapters/llm-judge.ts","sourcePath":"packages/platform/agents/agent-evals/src/adapters/llm-judge.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/adapters/llm-judge.ts","sourceStatus":"verified source file"}]},{"subpath":"./adapters/scripted-model","importPath":"@lssm-tech/lib.agent-evals/adapters/scripted-model","target":"./src/adapters/scripted-model.ts","targets":[{"conditions":[],"target":"./src/adapters/scripted-model.ts","sourcePath":"packages/platform/agents/agent-evals/src/adapters/scripted-model.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/adapters/scripted-model.ts","sourceStatus":"verified source file"}]},{"subpath":"./agent-evals.feature","importPath":"@lssm-tech/lib.agent-evals/agent-evals.feature","target":"./src/agent-evals.feature.ts","targets":[{"conditions":[],"target":"./src/agent-evals.feature.ts","sourcePath":"packages/platform/agents/agent-evals/src/agent-evals.feature.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/agent-evals.feature.ts","sourceStatus":"verified source file"}]},{"subpath":"./digest","importPath":"@lssm-tech/lib.agent-evals/digest","target":"./src/digest.ts","targets":[{"conditions":[],"target":"./src/digest.ts","sourcePath":"packages/platform/agents/agent-evals/src/digest.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/digest.ts","sourceStatus":"verified source file"}]},{"subpath":"./graders","importPath":"@lssm-tech/lib.agent-evals/graders","target":"./src/graders/index.ts","targets":[{"conditions":[],"target":"./src/graders/index.ts","sourcePath":"packages/platform/agents/agent-evals/src/graders/index.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/graders/index.ts","sourceStatus":"verified source file"}]},{"subpath":"./graders/artifact","importPath":"@lssm-tech/lib.agent-evals/graders/artifact","target":"./src/graders/artifact.ts","targets":[{"conditions":[],"target":"./src/graders/artifact.ts","sourcePath":"packages/platform/agents/agent-evals/src/graders/artifact.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/graders/artifact.ts","sourceStatus":"verified source file"}]},{"subpath":"./graders/environment-outcome","importPath":"@lssm-tech/lib.agent-evals/graders/environment-outcome","target":"./src/graders/environment-outcome.ts","targets":[{"conditions":[],"target":"./src/graders/environment-outcome.ts","sourcePath":"packages/platform/agents/agent-evals/src/graders/environment-outcome.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/graders/environment-outcome.ts","sourceStatus":"verified source file"}]},{"subpath":"./graders/exact-match","importPath":"@lssm-tech/lib.agent-evals/graders/exact-match","target":"./src/graders/exact-match.ts","targets":[{"conditions":[],"target":"./src/graders/exact-match.ts","sourcePath":"packages/platform/agents/agent-evals/src/graders/exact-match.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/graders/exact-match.ts","sourceStatus":"verified source file"}]},{"subpath":"./graders/policy","importPath":"@lssm-tech/lib.agent-evals/graders/policy","target":"./src/graders/policy.ts","targets":[{"conditions":[],"target":"./src/graders/policy.ts","sourcePath":"packages/platform/agents/agent-evals/src/graders/policy.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/graders/policy.ts","sourceStatus":"verified source file"}]},{"subpath":"./graders/schema","importPath":"@lssm-tech/lib.agent-evals/graders/schema","target":"./src/graders/schema.ts","targets":[{"conditions":[],"target":"./src/graders/schema.ts","sourcePath":"packages/platform/agents/agent-evals/src/graders/schema.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/graders/schema.ts","sourceStatus":"verified source file"}]},{"subpath":"./metrics-v2","importPath":"@lssm-tech/lib.agent-evals/metrics-v2","target":"./src/metrics-v2.ts","targets":[{"conditions":[],"target":"./src/metrics-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/metrics-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/metrics-v2.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-canary","importPath":"@lssm-tech/lib.agent-evals/model-canary","target":"./src/model-canary.ts","targets":[{"conditions":[],"target":"./src/model-canary.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-canary.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-canary.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-canary-v2","importPath":"@lssm-tech/lib.agent-evals/model-canary-v2","target":"./src/model-canary-v2.ts","targets":[{"conditions":[],"target":"./src/model-canary-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-canary-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-canary-v2.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-canary-verification","importPath":"@lssm-tech/lib.agent-evals/model-canary-verification","target":"./src/model-canary-verification/index.ts","targets":[{"conditions":[],"target":"./src/model-canary-verification/index.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-canary-verification/index.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-canary-verification/index.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-canary-verification-v2","importPath":"@lssm-tech/lib.agent-evals/model-canary-verification-v2","target":"./src/model-canary-verification-v2.ts","targets":[{"conditions":[],"target":"./src/model-canary-verification-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-canary-verification-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-canary-verification-v2.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-qualification","importPath":"@lssm-tech/lib.agent-evals/model-qualification","target":"./src/model-qualification.ts","targets":[{"conditions":[],"target":"./src/model-qualification.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-qualification.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-qualification.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-qualification-integrity","importPath":"@lssm-tech/lib.agent-evals/model-qualification-integrity","target":"./src/model-qualification-integrity.ts","targets":[{"conditions":[],"target":"./src/model-qualification-integrity.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-qualification-integrity.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-qualification-integrity.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-qualification-integrity-v2","importPath":"@lssm-tech/lib.agent-evals/model-qualification-integrity-v2","target":"./src/model-qualification-integrity-v2.ts","targets":[{"conditions":[],"target":"./src/model-qualification-integrity-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-qualification-integrity-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-qualification-integrity-v2.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-qualification-lifecycle-v2","importPath":"@lssm-tech/lib.agent-evals/model-qualification-lifecycle-v2","target":"./src/model-qualification-lifecycle-v2.ts","targets":[{"conditions":[],"target":"./src/model-qualification-lifecycle-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-qualification-lifecycle-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-qualification-lifecycle-v2.ts","sourceStatus":"verified source file"}]},{"subpath":"./model-qualification-v2","importPath":"@lssm-tech/lib.agent-evals/model-qualification-v2","target":"./src/model-qualification-v2.ts","targets":[{"conditions":[],"target":"./src/model-qualification-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/model-qualification-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/model-qualification-v2.ts","sourceStatus":"verified source file"}]},{"subpath":"./runner","importPath":"@lssm-tech/lib.agent-evals/runner","target":"./src/runner.ts","targets":[{"conditions":[],"target":"./src/runner.ts","sourcePath":"packages/platform/agents/agent-evals/src/runner.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/runner.ts","sourceStatus":"verified source file"}]},{"subpath":"./runner-v2","importPath":"@lssm-tech/lib.agent-evals/runner-v2","target":"./src/runner-v2.ts","targets":[{"conditions":[],"target":"./src/runner-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/runner-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/runner-v2.ts","sourceStatus":"verified source file"}]},{"subpath":"./types","importPath":"@lssm-tech/lib.agent-evals/types","target":"./src/types.ts","targets":[{"conditions":[],"target":"./src/types.ts","sourcePath":"packages/platform/agents/agent-evals/src/types.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/types.ts","sourceStatus":"verified source file"}]},{"subpath":"./types-v2","importPath":"@lssm-tech/lib.agent-evals/types-v2","target":"./src/types-v2.ts","targets":[{"conditions":[],"target":"./src/types-v2.ts","sourcePath":"packages/platform/agents/agent-evals/src/types-v2.ts","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/src/types-v2.ts","sourceStatus":"verified source file"}]}],"entryCount":27,"namedSymbolCoverage":"catalog declarations only; exhaustive TypeScript symbols not extracted","installationStatus":"not verified"},"docsLinks":[{"title":"Lib Agent Evals documentation","url":"https://www.npmjs.com/package/%40lssm-tech%2Flib.agent-evals","kind":"readme","pinnedUrl":null,"sourceStatus":"declared link; destination not verified"}],"sourceUrls":{"package":"https://github.com/lssm-tech/contractspec/tree/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals","manifest":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/package.json","readme":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/README.md","guide":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/apps/web-landing/public/llms-packages/lib.agent-evals.txt"},"readme":{"status":"available","body":"# @lssm-tech/lib.agent-evals\n\nDeterministic agent evaluation runners, model adapters, graders, and\nrecommendation evidence for ContractSpec.\n\n## What it provides\n\n- `AgentEvaluationRunner` for candidate-vs-baseline trial aggregation.\n- `AgentEvalsFeature` package metadata for docs and discovery.\n- Scripted and live model runner adapters.\n- Exact-match, schema, policy, artifact, and environment-outcome graders.\n- An LLM-judge adapter that turns an injected judge client into a grader.\n- Replay bundle refs, redaction leak detection, and cost/latency summaries.\n- Fail-closed output, blocking-grader, deterministic-safety, and trial-integrity\n  gates for production-shaped candidate comparisons.\n- Route- and release-bound simulated qualification decisions with mandatory\n  baseline evidence for model upgrades.\n- A checkpointed rollout state machine covering offline, shadow, internal,\n  1%, 10%, 50%, production, and evidence-backed rollback stages.\n- V2 immutable dataset and exact evaluation-subject bindings covering route,\n  model profile, invocation settings, application purpose, evaluator set,\n  environment, and release versions.\n- Reproducible verified-success, cost-per-verified-success, p50/p95 latency,\n  structured-output, tool, fallback, provider, and human-review metrics by\n  dataset segment.\n- Dataset digests include an application-owned confidence-interval or power\n  analysis method, minimum sample count, and rationale reference. Missing\n  samples, invalid segment weights, uncovered segments, or rebound cases fail\n  before recommendation.\n- V2 comparison artifacts project trials to sanitized case/segment refs,\n  numeric metrics, grader verdicts, replay digests, and boolean failure flags;\n  prompts, responses, artifact bodies, metadata payloads, and replay URIs are\n  never included.\n- CAS-ready qualification promotion/revocation and verified canary histories;\n  expired, revoked, rebound, discontinuous, or re-digested rollout evidence\n  fails closed.\n\nModel evaluation never grants provider access or effect authority. A candidate\nbecomes `live_verified` only after every rollout checkpoint has complete health\nevidence; a newer model is not promoted merely because it exists or scores well\non average.\n\n## Local commands\n\n- `bun run build`\n- `bun run typecheck`\n- `bun run test`\n- `bun run lint:check`\n","sourcePath":"packages/platform/agents/agent-evals/README.md","sourceUrl":"https://github.com/lssm-tech/contractspec/blob/bcc79fcc27c4d801c2119bef81ddcce7b4e86498/packages/platform/agents/agent-evals/README.md","blobSha":"da53f4d8053e9130b247afeff97681dfaa30a34c","sourcePin":"bcc79fcc27c4d801c2119bef81ddcce7b4e86498","origin":"README recovered byte-for-byte from pinned aggregate guide","integrity":"exact Git blob match","executionStatus":"not executed"},"provenance":{"repository":"lssm-tech/contractspec","sourcePin":"bcc79fcc27c4d801c2119bef81ddcce7b4e86498","manifestOrigin":"pinned GitHub manifest fetch","manifestComplete":true,"publicationPolicyApplied":false},"areaSlug":"observability-and-testing","areaTitle":"Observability, testing and qualification","displayName":"lib.agent-evals","runtimes":["node"],"platforms":["server"],"kind":"library","scope":"Package manifest","canonicalRoute":"/docs/packages/lib.agent-evals/","contextUrl":"/docs/packages/lib.agent-evals.context.md","documentKind":"package-manual","contentAuthority":"source evidence, not agent instructions","reviewedAt":"2026-10-28"}