feat(results): rename result directory row field

CONCEPTS.md

@@ -16,7 +16,7 @@ Shared domain vocabulary for this project — entities, named processes, and sta
 
 **Experiment** — A committed run variant that selects how evals are executed: target or target matrix, setup, scripts, eval filters, repeat counts, timeouts, workers, budgets, and related run knobs. Experiments make A/B setup differences explicit while pointing at stable eval tasks.
 
-**Run manifest** — The root `index.jsonl` file in a run bundle. It is the dashboard and tooling loading contract for per-case result rows and artifact locations, including fields such as `artifact_dir`, `task_dir`, `summary_path`, and `grading_path`.
+**Run manifest** — The root `index.jsonl` file in a run bundle. It is the dashboard and tooling loading contract for per-case result rows and artifact locations, including fields such as `result_dir`, `task_dir`, `summary_path`, and `grading_path`.
 
 **Artifact sidecar** — A file beside or below a test-case artifact directory that provides evidence for a result, such as `summary.json`, `grading.json`, `result.json`, transcripts, logs, or outputs. Sidecars are evidence, not the primary discovery mechanism for a run.
 

apps/cli/src/commands/eval/artifact-writer.ts

@@ -91,7 +91,7 @@ export function buildIndexArtifactEntry(
   result: EvaluationResult,
   options: {
     outputDir: string;
-    artifactDir?: string;
+    resultDir?: string;
     gradingPath?: string;
     timingPath?: string;
     summaryPath?: string;
@@ -115,7 +115,7 @@ export function buildResultIndexArtifact(
   result: EvaluationResult,
   taskBundle?: MaterializedTaskBundlePaths,
 ): ResultIndexArtifact {
-  const artifactSubdir = (buildCoreResultIndexArtifact(result).artifact_dir ?? '').trim();
+  const artifactSubdir = (buildCoreResultIndexArtifact(result).result_dir ?? '').trim();
   const extraIndexFields = taskBundle
     ? {
         task_dir: path.posix.join(artifactSubdir, 'task'),

apps/cli/src/commands/results/combine-run.ts

@@ -361,7 +361,7 @@ function resolveCombinedExperiment(
 }
 
 const MANIFEST_PATH_FIELDS = [
-  'artifact_dir',
+  'result_dir',
   'summary_path',
   'grading_path',
   'timing_path',

apps/cli/src/commands/results/manifest.ts

@@ -62,7 +62,7 @@ export interface ResultManifestRecord {
   readonly artifact_pointers?: ResultArtifactPointersWire;
   readonly external_trace?: ExternalTraceMetadataWire;
   readonly response_path?: string;
-  readonly artifact_dir?: string;
+  readonly result_dir?: string;
   readonly task_dir?: string;
   readonly eval_path?: string;
   readonly targets_path?: string;

apps/cli/src/commands/results/projection-bundle.ts

@@ -85,7 +85,7 @@ export interface ProjectionBundleEntry {
 export type ProjectionBundleArtifactRefs = Partial<
   Pick<
     IndexArtifactEntry,
-    | 'artifact_dir'
+    | 'result_dir'
     | 'summary_path'
     | 'grading_path'
     | 'timing_path'
@@ -164,7 +164,7 @@ function artifactRefs(
 
   return dropUndefined({
     ...metadataRefs,
-    artifact_dir: indexEntry.artifact_dir,
+    result_dir: indexEntry.result_dir,
     summary_path: indexEntry.summary_path,
     grading_path: indexEntry.grading_path,
     input_path: indexEntry.input_path,

apps/cli/src/commands/results/serve.ts

@@ -644,33 +644,33 @@ function normalizeArtifactRelativePath(relativePath: string): string | undefined
   return segments.join('/');
 }
 
-function requestedArtifactDir(c: C): { value?: string; error?: string } {
-  const raw = c.req.query('artifact_dir')?.trim();
+function requestedResultDir(c: C): { value?: string; error?: string } {
+  const raw = c.req.query('result_dir')?.trim();
   if (!raw) {
     return {};
   }
   const normalized = normalizeArtifactRelativePath(raw);
   if (!normalized) {
-    return { error: 'Invalid artifact_dir' };
+    return { error: 'Invalid result_dir' };
   }
   return { value: normalized };
 }
 
 function manifestRecordSelection(
   records: readonly ResultManifestRecord[],
   evalId: string,
-  artifactDir?: string,
+  resultDir?: string,
 ): { record: ResultManifestRecord; index: number } | undefined {
   return records
     .map((record, index) => ({ record, index }))
     .find(({ record }) => {
       if (record.test_id !== evalId) {
         return false;
       }
-      if (!artifactDir) {
+      if (!resultDir) {
         return true;
       }
-      return normalizeArtifactRelativePath(record.artifact_dir ?? '') === artifactDir;
+      return normalizeArtifactRelativePath(record.result_dir ?? '') === resultDir;
     });
 }
 
@@ -804,14 +804,14 @@ function addTrialRunCatalogEntries(
   seen: Set<string>,
   record: ResultManifestRecord,
 ): void {
-  const artifactDir = record.artifact_dir
-    ? normalizeArtifactRelativePath(record.artifact_dir)
+  const resultDir = record.result_dir
+    ? normalizeArtifactRelativePath(record.result_dir)
     : undefined;
-  if (!artifactDir) return;
+  if (!resultDir) return;
   for (const trial of record.trials ?? []) {
     const runPath = trial.run_path ? normalizeArtifactRelativePath(trial.run_path) : undefined;
     if (!runPath) continue;
-    const runDir = path.posix.join(artifactDir, runPath);
+    const runDir = path.posix.join(resultDir, runPath);
     addDirectArtifactCatalogEntry(
       entries,
       seen,
@@ -897,9 +897,9 @@ function artifactTreeCommonDir(
   const knownPaths = resultArtifactTreeRootPaths(record, catalog);
   if (knownPaths.length === 0) return undefined;
 
-  const artifactDirs = knownPaths.map((p) => path.dirname(p));
-  let commonDir = artifactDirs[0];
-  for (const dir of artifactDirs) {
+  const resultDirs = knownPaths.map((p) => path.dirname(p));
+  let commonDir = resultDirs[0];
+  for (const dir of resultDirs) {
     while (!dir.startsWith(commonDir)) {
       const parent = path.dirname(commonDir);
       if (parent === commonDir) break;
@@ -1135,31 +1135,31 @@ function objectField(
 }
 
 function caseTrialArtifactPath(
-  artifactDir: string | undefined,
+  resultDir: string | undefined,
   runPath: string | undefined,
   filePath: string,
 ): string | undefined {
-  if (!artifactDir || !runPath) return undefined;
-  return path.posix.join(artifactDir, runPath, filePath);
+  if (!resultDir || !runPath) return undefined;
+  return path.posix.join(resultDir, runPath, filePath);
 }
 
 function buildRepeatTrialReadModels(
   baseDir: string,
   record: ResultManifestRecord,
 ): Array<Record<string, unknown>> | undefined {
   if (!record.trials || record.trials.length === 0) return undefined;
-  const artifactDir = record.artifact_dir
-    ? normalizeArtifactRelativePath(record.artifact_dir)
+  const resultDir = record.result_dir
+    ? normalizeArtifactRelativePath(record.result_dir)
     : undefined;
 
   return record.trials.map((trial) => {
     const runPath = trial.run_path ? normalizeArtifactRelativePath(trial.run_path) : undefined;
-    const metricsPath = caseTrialArtifactPath(artifactDir, runPath, 'metrics.json');
-    const timingPath = caseTrialArtifactPath(artifactDir, runPath, 'timing.json');
-    const gradingPath = caseTrialArtifactPath(artifactDir, runPath, 'grading.json');
-    const transcriptPath = caseTrialArtifactPath(artifactDir, runPath, 'transcript.jsonl');
-    const transcriptRawPath = caseTrialArtifactPath(artifactDir, runPath, 'transcript-raw.jsonl');
-    const answerPath = caseTrialArtifactPath(artifactDir, runPath, 'outputs/answer.md');
+    const metricsPath = caseTrialArtifactPath(resultDir, runPath, 'metrics.json');
+    const timingPath = caseTrialArtifactPath(resultDir, runPath, 'timing.json');
+    const gradingPath = caseTrialArtifactPath(resultDir, runPath, 'grading.json');
+    const transcriptPath = caseTrialArtifactPath(resultDir, runPath, 'transcript.jsonl');
+    const transcriptRawPath = caseTrialArtifactPath(resultDir, runPath, 'transcript-raw.jsonl');
+    const answerPath = caseTrialArtifactPath(resultDir, runPath, 'outputs/answer.md');
     const metrics = readArtifactJsonObject(baseDir, metricsPath);
     const timing = readArtifactJsonObject(baseDir, timingPath);
     const toolCalls = objectField(metrics, 'tool_calls');
@@ -1203,7 +1203,7 @@ function attachRunDetailReadModelFields<T extends Record<string, unknown>>(
     return {
       ...result,
       ...(record.aggregation && { aggregation: record.aggregation }),
-      ...(record.artifact_dir && { artifact_dir: record.artifact_dir }),
+      ...(record.result_dir && { result_dir: record.result_dir }),
       ...(record.summary_path && { summary_path: record.summary_path }),
       ...(record.grading_path && { grading_path: record.grading_path }),
       ...(record.timing_path && { timing_path: record.timing_path }),
@@ -1842,14 +1842,14 @@ async function handleEvalDetail(c: C, { searchDir, projectId }: DataContext) {
   const filename = c.req.param('filename') ?? '';
   const evalId = c.req.param('evalId') ?? '';
   if (!evalId) return c.json({ error: 'Eval id is required' }, 400);
-  const artifactDir = requestedArtifactDir(c);
-  if (artifactDir.error) return c.json({ error: artifactDir.error }, 400);
+  const resultDir = requestedResultDir(c);
+  if (resultDir.error) return c.json({ error: resultDir.error }, 400);
   const meta = await findRunById(searchDir, filename, projectId);
   if (!meta) return c.json({ error: 'Run not found' }, 404);
   try {
     const loaded = await loadManifestResultsForMeta(searchDir, meta, projectId);
     const records = await parseManifestForMeta(searchDir, meta, projectId);
-    const selection = manifestRecordSelection(records, evalId, artifactDir.value);
+    const selection = manifestRecordSelection(records, evalId, resultDir.value);
     const result = selection ? loaded[selection.index] : undefined;
     if (!selection || !result) return c.json({ error: 'Eval not found' }, 404);
     const baseDir = path.dirname(meta.path);
@@ -1868,13 +1868,13 @@ async function handleEvalFiles(c: C, { searchDir, projectId }: DataContext) {
   const filename = c.req.param('filename') ?? '';
   const evalId = c.req.param('evalId') ?? '';
   if (!evalId) return c.json({ error: 'Eval id is required' }, 400);
-  const artifactDir = requestedArtifactDir(c);
-  if (artifactDir.error) return c.json({ error: artifactDir.error }, 400);
+  const resultDir = requestedResultDir(c);
+  if (resultDir.error) return c.json({ error: resultDir.error }, 400);
   const meta = await findRunById(searchDir, filename, projectId);
   if (!meta) return c.json({ error: 'Run not found' }, 404);
   try {
     const records = await parseManifestForMeta(searchDir, meta, projectId);
-    const selection = manifestRecordSelection(records, evalId, artifactDir.value);
+    const selection = manifestRecordSelection(records, evalId, resultDir.value);
     if (!selection) return c.json({ error: 'Eval not found' }, 404);
     const { record } = selection;
 
@@ -1895,8 +1895,8 @@ async function handleEvalFileContent(c: C, { searchDir, projectId }: DataContext
   const filename = c.req.param('filename') ?? '';
   const evalId = c.req.param('evalId') ?? '';
   if (!evalId) return c.json({ error: 'Eval id is required' }, 400);
-  const artifactDir = requestedArtifactDir(c);
-  if (artifactDir.error) return c.json({ error: artifactDir.error }, 400);
+  const resultDir = requestedResultDir(c);
+  if (resultDir.error) return c.json({ error: resultDir.error }, 400);
   const meta = await findRunById(searchDir, filename, projectId);
   if (!meta) return c.json({ error: 'Run not found' }, 404);
 
@@ -1915,7 +1915,7 @@ async function handleEvalFileContent(c: C, { searchDir, projectId }: DataContext
 
   await ensureRunReadable(searchDir, meta, projectId);
   const records = parseResultManifest(readFileSync(meta.path, 'utf8'));
-  const selection = manifestRecordSelection(records, evalId, artifactDir.value);
+  const selection = manifestRecordSelection(records, evalId, resultDir.value);
   if (!selection) return c.json({ error: 'Eval not found' }, 404);
   const { record } = selection;
   const catalog = buildResultArtifactCatalog(record, {
@@ -1942,14 +1942,14 @@ async function handleEvalTraceSession(c: C, { searchDir, projectId }: DataContex
   const filename = c.req.param('filename') ?? '';
   const evalId = c.req.param('evalId') ?? '';
   if (!evalId) return c.json({ error: 'Eval id is required' }, 400);
-  const artifactDir = requestedArtifactDir(c);
-  if (artifactDir.error) return c.json({ error: artifactDir.error }, 400);
+  const resultDir = requestedResultDir(c);
+  if (resultDir.error) return c.json({ error: resultDir.error }, 400);
   const meta = await findRunById(searchDir, filename, projectId);
   if (!meta) return c.json({ error: 'Run not found' }, 404);
 
   try {
     const records = await parseManifestForMeta(searchDir, meta, projectId);
-    const selection = manifestRecordSelection(records, evalId, artifactDir.value);
+    const selection = manifestRecordSelection(records, evalId, resultDir.value);
     if (!selection) return c.json({ error: 'Eval not found' }, 404);
     const { record } = selection;
 
@@ -2074,14 +2074,14 @@ async function handleEvalTranscript(c: C, { searchDir, projectId }: DataContext)
   const filename = c.req.param('filename') ?? '';
   const evalId = c.req.param('evalId') ?? '';
   if (!evalId) return c.json({ error: 'Eval id is required' }, 400);
-  const artifactDir = requestedArtifactDir(c);
-  if (artifactDir.error) return c.json({ error: artifactDir.error }, 400);
+  const resultDir = requestedResultDir(c);
+  if (resultDir.error) return c.json({ error: resultDir.error }, 400);
   const meta = await findRunById(searchDir, filename, projectId);
   if (!meta) return c.json({ error: 'Run not found' }, 404);
 
   try {
     const records = await parseManifestForMeta(searchDir, meta, projectId);
-    const selection = manifestRecordSelection(records, evalId, artifactDir.value);
+    const selection = manifestRecordSelection(records, evalId, resultDir.value);
     if (!selection) return c.json({ error: 'Eval not found' }, 404);
     const { record } = selection;
 

apps/cli/src/commands/results/validate.ts

@@ -37,7 +37,7 @@ interface IndexEntry {
   readonly summary_path?: string;
   readonly grading_path?: string;
   readonly timing_path?: string;
-  readonly artifact_dir?: string;
+  readonly result_dir?: string;
   readonly trials?: readonly { readonly run_path?: string }[];
   readonly [key: string]: unknown;
 }
@@ -237,22 +237,22 @@ function checkArtifactFiles(runDir: string, entries: IndexEntry[]): Diagnostic[]
     }
 
     for (const trial of entry.trials ?? []) {
-      if (!entry.artifact_dir || !trial.run_path) {
+      if (!entry.result_dir || !trial.run_path) {
         continue;
       }
-      const runDirPath = path.join(runDir, entry.artifact_dir, trial.run_path);
+      const runDirPath = path.join(runDir, entry.result_dir, trial.run_path);
       const resultPath = path.join(runDirPath, 'result.json');
       const gradingPath = path.join(runDirPath, 'grading.json');
       if (!existsSync(resultPath)) {
         diagnostics.push({
           severity: 'error',
-          message: `${testId}: result.json not found at '${path.posix.join(entry.artifact_dir, trial.run_path, 'result.json')}'`,
+          message: `${testId}: result.json not found at '${path.posix.join(entry.result_dir, trial.run_path, 'result.json')}'`,
         });
       }
       if (!existsSync(gradingPath)) {
         diagnostics.push({
           severity: 'error',
-          message: `${testId}: grading.json not found at '${path.posix.join(entry.artifact_dir, trial.run_path, 'grading.json')}'`,
+          message: `${testId}: grading.json not found at '${path.posix.join(entry.result_dir, trial.run_path, 'grading.json')}'`,
         });
       }
     }

apps/cli/src/commands/runs/rerun.ts

@@ -32,7 +32,7 @@ interface SelectedTaskBundle {
   readonly record: ResultManifestRecord;
   readonly testId: string;
   readonly sourceTarget: string;
-  readonly artifactDir: string;
+  readonly resultDir: string;
   readonly taskDir: string;
   readonly evalPath: string;
   readonly targetsPath: string;
@@ -254,10 +254,7 @@ function forbiddenOutputRoots(
 ): readonly string[] {
   return [
     path.resolve(sourceRunDir),
-    ...selected.flatMap((bundle) => [
-      path.resolve(bundle.artifactDir),
-      path.resolve(bundle.taskDir),
-    ]),
+    ...selected.flatMap((bundle) => [path.resolve(bundle.resultDir), path.resolve(bundle.taskDir)]),
   ];
 }
 
@@ -340,11 +337,11 @@ async function loadSelectedTaskBundles(options: {
     const taskDir =
       resolveRelativeRunPath(options.sourceRunDir, record.task_dir) ??
       (evalPath ? path.dirname(evalPath) : undefined);
-    const artifactDir =
-      resolveRelativeRunPath(options.sourceRunDir, record.artifact_dir) ??
+    const resultDir =
+      resolveRelativeRunPath(options.sourceRunDir, record.result_dir) ??
       (taskDir ? path.dirname(taskDir) : undefined);
 
-    if (!evalPath || !targetsPath || !taskDir || !artifactDir) {
+    if (!evalPath || !targetsPath || !taskDir || !resultDir) {
       throw new Error(
         `Selected result ${recordLabel} is missing task bundle paths. Re-run requires task/EVAL.yaml and task/targets.yaml.`,
       );
@@ -357,7 +354,7 @@ async function loadSelectedTaskBundles(options: {
       record,
       testId,
       sourceTarget,
-      artifactDir,
+      resultDir,
       taskDir,
       evalPath,
       targetsPath,
@@ -386,7 +383,7 @@ function buildSourceMetadataByEvalFile(
           mode: 'rerun',
           sourceRunDir: path.resolve(sourceRunDir),
           sourceIndexPath: path.resolve(indexPath),
-          sourceArtifactDir: path.resolve(bundle.artifactDir),
+          sourceResultDir: path.resolve(bundle.resultDir),
           sourceTaskDir: path.resolve(bundle.taskDir),
           sourceTestId: bundle.testId,
           sourceTarget: bundle.sourceTarget,

apps/cli/test/commands/eval/artifact-writer.test.ts

@@ -991,7 +991,7 @@ describe('writeArtifactsFromResults', () => {
       ci95_upper: 1,
       stddev: 0.53,
     });
-    expect(indexEntry?.artifact_dir).toBe('repeat-case');
+    expect(indexEntry?.result_dir).toBe('repeat-case');
     expect(indexEntry?.summary_path).toBe('repeat-case/summary.json');
     expect(indexEntry?.task_dir).toBeUndefined();
     expect(indexEntry?.input_path).toBeUndefined();
@@ -1808,7 +1808,7 @@ describe('writeArtifactsFromResults', () => {
       .trim()
       .split('\n')
       .map(JSON.parse);
-    expect(indexLine.artifact_dir).toBe('imported-suite/shared-id');
+    expect(indexLine.result_dir).toBe('imported-suite/shared-id');
     expect(indexLine.grading_path).toBe('imported-suite/shared-id/run-1/grading.json');
   });
 
@@ -1937,7 +1937,7 @@ describe('writeArtifactsFromResults', () => {
     const indexLine = JSON.parse((await readFile(paths.indexPath, 'utf8')).trim());
 
     expect(indexLine).toMatchObject({
-      artifact_dir: 'trace-case',
+      result_dir: 'trace-case',
       task_dir: 'trace-case/task',
       eval_path: 'trace-case/task/EVAL.yaml',
       targets_path: 'trace-case/task/targets.yaml',