Functions — github.com/anomalyco/opencode-bench

hub / github.com/anomalyco/opencode-bench / functions

Functions110 in github.com/anomalyco/opencode-bench

⨍ Functions 110◇ Types & classes 15

↓ 9 callersFunctiontestConsistency

( judge: Judge, systemPrompt: string, diffPair: DiffPair, scoreType: "logic-equivalence" | "api-signat

tests/judgeConsistencyMedium.test.ts:74

↓ 7 callersFunctionaverage

(values: number[])

src/util/math.ts:1

↓ 6 callersFunctioncloneScores

( scores: EvaluationRunExport["scores"], averageShift: number, varianceShift = 0, )

scripts/discord-sample.ts:73

↓ 6 callersFunctioncreateJudge

(name: JudgeName)

tests/scoreAggregation.test.ts:20

↓ 6 callersFunctionevaluateAndShow

( name: string, systemPrompt: string, reference: string, candidate: string, scoreType: string, )

scripts/show-test-outputs.ts:16

↓ 6 callersFunctiontestConsistency

* Run consistency test: evaluate the same diff pair N times and check results.

tests/judgeConsistency.test.ts:81

↓ 4 callersFunctiongetZenLanguageModel

(modelId: string)

src/zenModels.ts:111

↓ 4 callersFunctionweightedSum

(entries: { value: number; weight: number }[])

src/util/math.ts:11

↓ 3 callersFunctioncalculateMean

(values: number[])

src/util/statistics.ts:8

↓ 3 callersFunctiondate

()

src/util/logger.ts:19

↓ 3 callersFunctionformat

(...messages: any[])

src/util/logger.ts:7

↓ 3 callersFunctionrunCommands

( commands: string[], opts: { logger: Logger.Instance; cwd: string }, )

src/eval.ts:281

↓ 3 callersFunctiontestConsistency

( judge: Judge, systemPrompt: string, diffPair: DiffPair, scoreType: "logic-equivalence", runs: numb

tests/judgeConsistencyComplex.test.ts:57

↓ 2 callersFunctionbuildBarChartUrl

( data: BarChartData, options?: { width?: number; height?: number; backgroundColor?: string;

src/util/charts.ts:135

↓ 2 callersFunctionfileExists

(path: string)

src/util/fs.ts:4

↓ 2 callersFunctionformatExecution

(execution: Metric.CommandExecution)

src/metrics/checks.ts:115

↓ 2 callersFunctionformatUsage

(usage: { input: number; output: number })

cli.ts:69

↓ 2 callersFunctiongetRequestClient

()

src/util/github.ts:7

↓ 2 callersFunctionlistNames

()

src/tasks/index.ts:74

↓ 2 callersFunctionparseArgs

(argv: string[])

scripts/determine-job-url.ts:26

↓ 2 callersFunctionsessionKey

(model: string, cwd: string)

src/agents/codex.ts:27

↓ 2 callersFunctiontoInputId

(agent: string, model: string)

scripts/sync-workflow-inputs.ts:25

↓ 2 callersFunctionusage

()

scripts/determine-job-url.ts:18

↓ 1 callersFunctionbuildAverageChartUrl

( evalName: string, models: ModelSummary[], )

scripts/discord-sample.ts:372

↓ 1 callersFunctionbuildDynamicContext

(runs: EvaluationRunExport[])

scripts/analysis.ts:38

↓ 1 callersFunctionbuildInputMapping

()

scripts/build-workflow-matrix.ts:42

↓ 1 callersFunctionbuildOverallChartUrl

( evalSummaries: EvalSummary[], )

scripts/discord-sample.ts:397

↓ 1 callersFunctionbuildPayloads

( evalSummaries: EvalSummary[], analysisLinks: Map<string, string>, )

scripts/discord-sample.ts:463

↓ 1 callersFunctionbuildRadarChartUrl

( data: RadarChartData, options?: { width?: number; height?: number; backgroundColor?: string;

src/util/charts.ts:27

↓ 1 callersFunctioncalculateStdDev

(values: number[])

src/util/statistics.ts:19

↓ 1 callersFunctioncleanupRepository

( cwd: string, logger: Logger.Instance, )

src/eval.ts:380

↓ 1 callersFunctioncloneRepositoryAtCommit

(repo: string, commitSha: string)

src/eval.ts:372

↓ 1 callersFunctioncollectJsonFiles

(root: string)

scripts/merge-benchmark-exports.ts:8

↓ 1 callersFunctioncreateDiffComparisonPrompt

( diffPair: DiffPair, scoreType: "logic-equivalence", )

tests/judgeConsistencyComplex.test.ts:49

↓ 1 callersFunctioncreateDiffComparisonPrompt

* Helper to create the user prompt for diff comparison. * Uses the same prompt creation functions as production.

tests/judgeConsistency.test.ts:65

↓ 1 callersFunctioncreateDiffComparisonPrompt

( diffPair: DiffPair, scoreType: "logic-equivalence" | "api-signature" | "integration-points", )

tests/judgeConsistencyMedium.test.ts:59

↓ 1 callersFunctioncreateRegistration

( name: string, module: { default?: Definition<TModel>; models?: ReadonlyArray<TModel>;

src/agents/index.ts:59

↓ 1 callersFunctionensureProviders

()

src/zenModels.ts:51

↓ 1 callersFunctionevaluateWithJudge

( judge: Judge, systemPrompt: string, userPrompt: string, )

tests/judgeCompatibility.test.ts:50

↓ 1 callersFunctionevaluateWithJudge

* High Complexity Judge Consistency Tests * * These tests verify that judges handle very complex scenarios: * - Architectural refactors (class → fu

tests/judgeConsistencyComplex.test.ts:28

↓ 1 callersFunctionevaluateWithJudge

* Evaluate a diff pair directly using the judge's LLM. * This bypasses git operations and uses static diffs for consistency testing.

tests/judgeConsistency.test.ts:40

↓ 1 callersFunctionevaluateWithJudge

* Medium Complexity Judge Consistency Tests * * These tests verify that judges produce consistent scores on more complex scenarios: * - Nested loop

tests/judgeConsistencyMedium.test.ts:38

↓ 1 callersFunctionfetchCommits

( owner: string, repo: string, from: string, to: string, )

src/util/github.ts:55

↓ 1 callersFunctionfetchComparisonDiff

( owner: string, repo: string, from: string, to: string, )

src/util/github.ts:18

↓ 1 callersFunctionfetchJobs

( owner: string, repo: string, runId: number, )

scripts/determine-job-url.ts:49

↓ 1 callersFunctionfinalizeChanges

(baselineCommit: string)

src/eval.ts:168

↓ 1 callersFunctionformatExecutionForLog

(execution: Metric.CommandExecution)

src/eval.ts:358

↓ 1 callersFunctionformatScore

(score: Eval.Result["score"])

cli.ts:71

↓ 1 callersFunctiongenerateAnalysis

(runs: EvaluationRunExport[])

scripts/analysis.ts:98

↓ 1 callersFunctiongenerateDiff

(baselineCommit: string)

src/eval.ts:224

↓ 1 callersFunctiongeneratePrompt

( def: z.infer<typeof definitionSchema>, diff: CommitDiff, opts: { logger: Logger.Instance }, )

src/tasks/index.ts:176

↓ 1 callersFunctiongetOrCreateThread

(model: string, cwd: string)

src/agents/codex.ts:54

↓ 1 callersFunctioninferEndpoint

(modelId: string)

src/zenModels.ts:89

↓ 1 callersFunctionjudgeScore

( criterionName: string, judge: string, context: Metric.Context, opts: { logger: Logger.Instan

src/eval.ts:240

↓ 1 callersFunctionload

()

src/tasks/index.ts:85

↓ 1 callersFunctionloadAnalysisLinks

()

scripts/discord-sample.ts:434

↓ 1 callersFunctionloadDataset

()

scripts/build-workflow-matrix.ts:73

↓ 1 callersFunctionloadExport

()

scripts/discord-sample.ts:256

↓ 1 callersFunctionload_runs

(files: List[Path])

visualize_instability.py:24

↓ 1 callersFunctionlogError

(value: unknown, options: Agent.RunOptions)

src/agents/claude-code.ts:37

↓ 1 callersFunctionlogJson

(value: unknown, options: Agent.RunOptions)

src/agents/claude-code.ts:23

↓ 1 callersFunctionlogTurnItems

(items: ThreadItem[], options: Agent.RunOptions)

src/agents/codex.ts:31

↓ 1 callersFunctionmain

()

visualize_instability.py:135

↓ 1 callersFunctionmain

()

scripts/build-workflow-matrix.ts:88

↓ 1 callersFunctionmain

()

scripts/determine-job-url.ts:97

↓ 1 callersFunctionmain

()

scripts/analysis.ts:118

↓ 1 callersFunctionmain

()

scripts/discord-sample.ts:599

↓ 1 callersFunctionmain

()

scripts/sync-workflow-inputs.ts:34

↓ 1 callersFunctionmelt_scores

(runs: List[Dict])

visualize_instability.py:45

↓ 1 callersFunctionnormalizeModelId

(modelId: string)

src/zenModels.ts:78

↓ 1 callersFunctionplot_instability

(df: pd.DataFrame, output: Path)

visualize_instability.py:61

↓ 1 callersFunctionresolveContentLink

()

scripts/discord-sample.ts:542

↓ 1 callersFunctionresolveZenApiKey

()

src/zenModels.ts:24

↓ 1 callersFunctionresolveZenBaseUrl

()

src/zenModels.ts:42

↓ 1 callersFunctionrunCommand

(command: string, cwd: string)

src/eval.ts:297

↓ 1 callersFunctionrunOnce

( agentName: string, modelId: string, taskId: string, opts: { logger: Logger.Instance;

src/eval.ts:41

↓ 1 callersFunctionsendWebhook

( webhookUrl: string, payload: unknown, partIndex: number, totalParts: number, )

scripts/discord-sample.ts:567

↓ 1 callersFunctionserializeError

(error: unknown)

src/agents/claude-code.ts:51

↓ 1 callersFunctionsessionKey

(model: string, cwd: string)

src/agents/claude-code.ts:19

↓ 1 callersFunctionsessionKey

(model: string, cwd: string)

src/agents/opencode.ts:53

↓ 1 callersFunctiontoDescription

(agent: string, model: string)

scripts/sync-workflow-inputs.ts:30

↓ 1 callersFunctiontoEvalSummaries

(exportData: EvaluationRunExport[])

scripts/discord-sample.ts:267

↓ 1 callersFunctionusage

()

scripts/analysis.ts:110

↓ 1 callersFunctionvariance

(avg: number, values: number[])

src/util/math.ts:5

↓ 1 callersFunctionwithRetries

( fn: () => Promise<T>, options: { retries: number; timeoutMs: number; logger: Logger.Instance

src/util/retry.ts:3

FunctionbuildModelRadarChartUrl

* Builds a shareable QuickChart radar chart URL for a model's per-metric scores. * Currently unused in the Discord webhook, but retained for future s

scripts/discord-sample.ts:343

FunctioncalculateConfidenceInterval

( values: number[], _confidenceLevel: number = 0.95, )

src/util/statistics.ts:33

FunctioncalculateFleissKappa

(judgeScores: number[][])

src/util/statistics.ts:80

FunctioncalculateRange

(values: number[])

src/util/statistics.ts:143

Functioncleanup

()

src/agents/opencode.ts:136

FunctioncomputeAverageScore

(model: ModelSummary)

scripts/discord-sample.ts:362

Functioncreate

(prefix?: string)

src/util/logger.ts:6

FunctioncreateUserPrompt

(context: Metric.Context)

src/metrics/test-coverage.ts:163

FunctioncreateUserPrompt

(context: Metric.Context)

src/metrics/logic-equivalence.ts:210

FunctioncreateUserPrompt

(context: Metric.Context)

src/metrics/api-signature.ts:135

FunctioncreateUserPrompt

(context: Metric.Context)

src/metrics/integration-points.ts:166

FunctioncreateUserPrompt

(context: Metric.Context)

src/metrics/checks.ts:93

FunctionformatFallbackSummary

(runs: EvaluationRunExport[])

scripts/analysis.ts:65

Functiongenerate

(opts: { logger: Logger.Instance })

src/tasks/index.ts:105

Functionget

(taskId: string)

src/tasks/index.ts:57

next →1–100 of 110, ranked by callers