@iowarp/clio-coder 0.3.9 → 0.4.1

package/dist/{eval-IJ5VEZDJ.js → eval-GXLL44RD.js}

@@ -1,72 +1,82 @@
 import { createRequire as __clioCreateRequire } from "node:module"; const require = __clioCreateRequire(import.meta.url);
 import {
-  discoverAgentRecipes
-} from "./chunk-HCBCAYZU.js";
-import "./chunk-AD7Y7STJ.js";
-import "./chunk-AMKHQW3C.js";
+  EVAL_SUITE_V2_VERSION
+} from "./chunk-RKSR6VSF.js";
 import {
   loadFragments,
   renderCodewikiDigest
-} from "./chunk-47CMYGET.js";
+} from "./chunk-3QSOM6PA.js";
 import {
-  EVAL_BEHAVIOR_METRIC_DEFINITIONS_V1,
-  EVAL_EXECUTION_ENVELOPE_SCHEMA_V1,
-  EVAL_TRACKED_METRIC_NAMES,
-  EVAL_VERDICT_SCHEMA_V1,
   EvalTaskFileError,
   TRUST_STATUS_AXES,
   adaptRunReceiptTrustStatus,
   assertComparableTrackedMetricSources,
-  assertEvalBehaviorReferencesVerdictV1,
-  buildEvalBehaviorMetricsV1,
-  createEvalId,
   evalClioProvenance,
   evalEnvironmentProvenance,
-  evalHarnessMetricsFromReceipt,
   evalServingConfiguration,
   evalServingObservationFrom,
   formatTrustSummary,
   inspectRunReceiptTrustStatus,
+  loadEvalTaskFile,
+  summarizeTrustStatus,
+  verifyReceiptIntegrity
+} from "./chunk-6NJQITNH.js";
+import {
+  listSessionLedgerRefs,
+  parseSessionEntries
+} from "./chunk-TVH4ONAM.js";
+import "./chunk-3DPEIQKN.js";
+import {
+  EVAL_BEHAVIOR_METRIC_DEFINITIONS_V1,
+  EVAL_EXECUTION_ENVELOPE_SCHEMA_V1,
+  EVAL_TRACKED_METRIC_NAMES,
+  EVAL_VERDICT_SCHEMA_V1,
+  assertEvalBehaviorReferencesVerdictV1,
+  buildEvalBehaviorMetricsV1,
+  createEvalId,
+  evalHarnessMetricsFromReceipt,
+  evalServingConfigurationOf,
   judgeEvalBehaviorV1,
   loadEvalArtifactV4,
-  loadEvalTaskFile,
+  normalizeEvalRunnerKind,
+  normalizeEvalSchemaId,
   parseEvalBehaviorScenarioV1,
   parseEvalExecutionMatrixDimensionsV1,
   parseEvalVerdictEnvelopeV1,
   renderEvalServingConfiguration,
   sameEvalServingConfiguration,
-  summarizeTrustStatus,
-  verifyReceiptIntegrity,
   writeEvalArtifactV4
-} from "./chunk-MLOK6ZOS.js";
-import {
-  listSessionLedgerRefs,
-  parseSessionEntries
-} from "./chunk-LXPJXFM5.js";
-import "./chunk-3DPEIQKN.js";
+} from "./chunk-6CCS4G3W.js";
 import "./chunk-W6GROXXM.js";
-import "./chunk-RKKLTLYB.js";
-import "./chunk-5QKCQQ3E.js";
 import {
   shellQuote
 } from "./chunk-TXOTCRLG.js";
 import "./chunk-HVDIIIQW.js";
-import "./chunk-VPTUJU4P.js";
-import "./chunk-2JDWVJND.js";
-import "./chunk-H7IXIC72.js";
+import {
+  discoverAgentRecipes
+} from "./chunk-ZJLUDYFY.js";
+import "./chunk-HUAS7ITX.js";
+import "./chunk-NDDGJUXD.js";
+import "./chunk-FEFIFZTL.js";
+import "./chunk-RKKLTLYB.js";
+import "./chunk-SUW5DORT.js";
+import "./chunk-7EPLI7VL.js";
+import "./chunk-6DWBAZ5U.js";
 import {
   createSafetyPolicyEngine
-} from "./chunk-N3PBVRTZ.js";
-import "./chunk-A2NJGIB3.js";
+} from "./chunk-IXJT6DCX.js";
+import "./chunk-H7IXIC72.js";
+import "./chunk-VN3SHNBN.js";
+import "./chunk-FSP7CMNU.js";
+import "./chunk-HHV2GANA.js";
 import {
   agentSpecFingerprint,
   normalizeAgentSpec
-} from "./chunk-S4COXYBG.js";
-import "./chunk-MV3K5QF2.js";
-import "./chunk-RAPCMZL4.js";
+} from "./chunk-IG7BCQBA.js";
+import "./chunk-GCSMB2KY.js";
 import "./chunk-UL3WSD3F.js";
 import "./chunk-ECH6PKUQ.js";
-import "./chunk-CGKSTWHD.js";
+import "./chunk-K6BSR66V.js";
 import {
   readCodewiki,
   structuralCodewikiHash
@@ -79,30 +89,39 @@ import {
 } from "./chunk-33YXPOE3.js";
 import "./chunk-7CR24IG7.js";
 import "./chunk-XPLRXC72.js";
-import "./chunk-IFBNV6H6.js";
+import "./chunk-ZW55JB7N.js";
 import {
   printError
-} from "./chunk-XK56QHLX.js";
+} from "./chunk-VPKWYKEY.js";
 import "./chunk-5TSRNF4G.js";
 import "./chunk-CFGTUFWB.js";
 import {
   extractReasoningTokens
-} from "./chunk-AEYBF3TB.js";
-import "./chunk-IHXBNWMM.js";
-import "./chunk-B5CSFE7B.js";
-import "./chunk-PNY46YEY.js";
-import "./chunk-FQ4SKYE4.js";
-import "./chunk-RKRLDWD3.js";
+} from "./chunk-CWVRRIEI.js";
+import "./chunk-MQXIVJ35.js";
 import {
   InvalidIdError
-} from "./chunk-KV2AOLDF.js";
+} from "./chunk-R346GLFC.js";
+import "./chunk-HLW2MRKE.js";
+import {
+  normalizeClioCoderEventRecord,
+  normalizeClioCoderEventType
+} from "./chunk-BTK3SSSK.js";
+import "./chunk-IHXBNWMM.js";
+import "./chunk-EIMVLWB3.js";
+import {
+  namingCompatibilityEnvironment
+} from "./chunk-D7E7WCDF.js";
+import "./chunk-HJWWJ6IL.js";
+import "./chunk-JIEGK6UF.js";
+import "./chunk-EBY63B7G.js";
 import "./chunk-6EJMN2Y3.js";
 import "./chunk-IWHMRKLL.js";
-import "./chunk-XDOQXGFO.js";
+import "./chunk-BWW4HLO4.js";
 import "./chunk-LL4KHSZI.js";
-import "./chunk-4ZG3XFUR.js";
-import "./chunk-EQ63NRB7.js";
-import "./chunk-SST6Z5JA.js";
+import "./chunk-NLFAQR7Z.js";
+import "./chunk-WNP7O5WZ.js";
+import "./chunk-336NSSKI.js";
 import "./chunk-IKCO5N3L.js";
 import "./chunk-3I7MS7N2.js";
 import {
@@ -111,7 +130,7 @@ import {
 import {
   clioDataDir,
   clioStateDir
-} from "./chunk-BNAZZHFG.js";
+} from "./chunk-BYP5D4HI.js";
 import "./chunk-WEPFGWHJ.js";
 import "./chunk-YXLYO42X.js";
 import {
@@ -453,8 +472,8 @@ var EvalServingConfigurationDriftError = class extends Error {
 function compareEvalArtifactsV4(baseline, candidate, options = {}) {
   const baselineTokens = baseline.summary.tokens;
   const candidateTokens = candidate.summary.tokens;
-  const baselineServing = servingConfigurationOf(baseline);
-  const candidateServing = servingConfigurationOf(candidate);
+  const baselineServing = evalServingConfigurationOf(baseline);
+  const candidateServing = evalServingConfigurationOf(candidate);
   const configDrift = !sameEvalServingConfiguration(baselineServing, candidateServing);
   if (configDrift && options.allowConfigDrift !== true) {
     throw new EvalServingConfigurationDriftError(baselineServing, candidateServing);
@@ -654,17 +673,6 @@ function metricComparison(scenarioId, metric, baseline, candidate) {
     varianceChange: classifyChange(baseline.variance ?? null, candidate.variance ?? null, "lower")
   };
 }
-function servingConfigurationOf(artifact) {
-  return artifact.servingConfiguration ?? {
-    targetId: artifact.matrix.target,
-    runtimeId: null,
-    modelId: artifact.matrix.model,
-    serverBuild: null,
-    total_slots: null,
-    thinkingLevel: artifact.matrix.thinking,
-    compiledPromptHash: null
-  };
-}
 function normalizeMetricFilter(metric) {
   if (metric === void 0) return void 0;
   const trimmed = metric.trim();
@@ -1052,13 +1060,7 @@ import { dirname, resolve } from "node:path";
 
 // src/domains/eval/schema/validate.ts
 init_esm_shims();
-
-// src/domains/eval/schema/suite.ts
-init_esm_shims();
-var EVAL_SUITE_V2_VERSION = 2;
-
-// src/domains/eval/schema/validate.ts
-var RUNNER_KINDS = /* @__PURE__ */ new Set(["clio-run", "context-index", "context-init", "external-command"]);
+var RUNNER_KINDS = /* @__PURE__ */ new Set(["clio-coder-run", "context-index", "context-init", "external-command"]);
 var WORKSPACE_KINDS = /* @__PURE__ */ new Set(["local", "git", "temp-copy"]);
 var OPS = /* @__PURE__ */ new Set(["lt", "lte", "gt", "gte", "eq", "neq"]);
 function validateEvalSuiteV2(value) {
@@ -1222,9 +1224,12 @@ function readRunner(value, path, issues) {
     issues.push({ path, message: "expected object" });
     return null;
   }
-  const kind = value.kind;
+  const kind = normalizeEvalRunnerKind(value.kind);
   if (typeof kind !== "string" || !RUNNER_KINDS.has(kind)) {
-    issues.push({ path: `${path}.kind`, message: "expected clio-run, context-index, context-init, or external-command" });
+    issues.push({
+      path: `${path}.kind`,
+      message: "expected clio-coder-run, context-index, context-init, or external-command"
+    });
     return null;
   }
   const prompt = optionalString(value, "prompt");
@@ -1233,8 +1238,8 @@ function readRunner(value, path, issues) {
   if (autonomy !== void 0 && !["read-only", "suggest", "auto-edit", "full-auto"].includes(autonomy)) {
     issues.push({ path: `${path}.autonomy`, message: "expected read-only, suggest, auto-edit, or full-auto" });
   }
-  if (agent !== void 0 && kind !== "clio-run") {
-    issues.push({ path: `${path}.agent`, message: "agent is only valid on the clio-run runner" });
+  if (agent !== void 0 && kind !== "clio-coder-run") {
+    issues.push({ path: `${path}.agent`, message: "agent is only valid on the clio-coder-run runner" });
   }
   const command = optionalString(value, "command");
   return {
@@ -1535,7 +1540,7 @@ function promptFragmentIdentities(manifest, recipe, autonomy) {
     })).sort((left, right) => left.id.localeCompare(right.id));
   }
   if (recipe === null) return [];
-  const selected = ["identity.clio-worker", "operating.contract", "operating.worker"];
+  const selected = ["identity.clio-coder-worker", "operating.contract", "operating.worker"];
   if (autonomy !== null) selected.push(`safety.${autonomy}`);
   const fragments = [];
   try {
@@ -3039,7 +3044,7 @@ function createJsonlMetricCapture() {
   };
 }
 function compactMetricEvent(event) {
-  const type = event.type;
+  const type = typeof event.type === "string" ? normalizeClioCoderEventType(event.type) : event.type;
   if (type === "tool_execution_start") {
     const toolName = stringField(event, "toolName");
     if (toolName !== "dispatch" && toolName !== "code_nav" && toolName !== "read" && toolName !== "grep") {
@@ -3061,7 +3066,7 @@ function compactMetricEvent(event) {
       ...stringField(event, "outcome") !== void 0 ? { outcome: stringField(event, "outcome") } : {}
     };
   }
-  if (type !== "clio_tool_finish" || !isRecord8(event.payload)) return null;
+  if (type !== "clio_coder_tool_finish" || !isRecord8(event.payload)) return null;
   return {
     type,
     payload: {
@@ -3150,7 +3155,7 @@ function toolCallMetricsFromJsonl(stdout) {
     try {
       const parsed = JSON.parse(line);
       if (!isRecord9(parsed)) continue;
-      event = parsed;
+      event = normalizeClioCoderEventRecord(parsed);
     } catch {
       continue;
     }
@@ -3163,7 +3168,7 @@ function toolCallMetricsFromJsonl(stdout) {
       recordToolOutcome(executionEnds, toolOutcome(event) ?? (event.isError === true ? "error" : "ok"));
       continue;
     }
-    if (event.type !== "clio_tool_finish" || !isRecord9(event.payload)) continue;
+    if (event.type !== "clio_coder_tool_finish" || !isRecord9(event.payload)) continue;
     const outcome = toolOutcome(event.payload);
     if (outcome === void 0) continue;
     const callId = stringField2(event.payload, "toolCallId") ?? stringField2(event, "toolCallId");
@@ -3188,7 +3193,7 @@ function toolBehaviorMetricEntriesFromJsonl(stdout, cwd, readObservation) {
     try {
       const parsed = JSON.parse(line);
       if (!isRecord9(parsed)) continue;
-      event = parsed;
+      event = normalizeClioCoderEventRecord(parsed);
     } catch {
       continue;
     }
@@ -3210,7 +3215,7 @@ function toolBehaviorMetricEntriesFromJsonl(stdout, cwd, readObservation) {
       executionEnds.push({ callId: callId2, tool: tool2, outcome: toolOutcome(event) ?? (event.isError === true ? "error" : "ok") });
       continue;
     }
-    if (event.type !== "clio_tool_finish" || !isRecord9(event.payload)) continue;
+    if (event.type !== "clio_coder_tool_finish" || !isRecord9(event.payload)) continue;
     const outcome = toolOutcome(event.payload);
     const tool = stringField2(event.payload, "tool");
     if (outcome === void 0 || tool === void 0) continue;
@@ -3609,7 +3614,7 @@ import { tmpdir } from "node:os";
 import { resolve as resolve4 } from "node:path";
 async function prepareGitWorkspace(workspace) {
   if (workspace.url === void 0) throw new Error("git workspace requires url");
-  const dest = await mkdtemp(resolve4(tmpdir(), "clio-eval-git-"));
+  const dest = await mkdtemp(resolve4(tmpdir(), "clio-coder-eval-git-"));
   try {
     await runGit(["clone", "--quiet", workspace.url, dest], process.cwd());
     const ref = workspace.checkout ?? workspace.commit;
@@ -3663,7 +3668,7 @@ var execFileAsync = promisify(execFile);
 var GIT_FILE_LIST_LIMIT_BYTES = 128 * 1024 * 1024;
 async function prepareTempCopyWorkspace(baseDir, workspace, options = {}) {
   const source = resolve6(baseDir, workspace.path ?? ".");
-  const dest = await mkdtemp2(resolve6(options.tempRoot ?? tmpdir2(), "clio-eval-workspace-"));
+  const dest = await mkdtemp2(resolve6(options.tempRoot ?? tmpdir2(), "clio-coder-eval-workspace-"));
   try {
     const selection = await gitCopySelection(source);
     const excludes = workspace.excludes ?? [];
@@ -3845,7 +3850,7 @@ async function runMatrixItem(loaded, task, target, repeatIndex, clioEntry, fresh
   let receipt = null;
   let runnerWallTimeMs = 0;
   let executionObservation;
-  const stateDir = await mkdtemp3(resolve7(tempCopy?.tempRoot ?? tmpdir3(), "clio-eval-state-"));
+  const stateDir = await mkdtemp3(resolve7(tempCopy?.tempRoot ?? tmpdir3(), "clio-coder-eval-state-"));
   try {
     workspace = await prepareWorkspace(loaded.baseDir, task, freshWorkspace, tempCopy);
     const setup = await runCommandVerifiers(task.workspace.setup ?? [], workspace.dir, task.timeoutMs);
@@ -3862,7 +3867,11 @@ async function runMatrixItem(loaded, task, target, repeatIndex, clioEntry, fresh
     const receiptExitCode = runner.exitCode;
     const journalMetrics = invariantMetrics(stateDir, receiptExitCode);
     const measurement = await measureTaskOutcome(task, workspace.dir, {
-      CLIO_EVAL_RUNNER_STDOUT_FILE: runnerStdoutFile
+      ...namingCompatibilityEnvironment(
+        "CLIO_CODER_EVAL_RUNNER_STDOUT_FILE",
+        "CLIO_EVAL_RUNNER_STDOUT_FILE",
+        runnerStdoutFile
+      )
     });
     executionObservation = measurement.executionObservation;
     const metrics = {
@@ -4043,13 +4052,13 @@ function graderBehaviorMeasurement(stdout) {
       continue;
     }
     if (!isRecord10(value)) continue;
-    if (value.schema === "clio.eval.measure.v1" && isRecord10(value.metrics)) {
+    if (normalizeEvalSchemaId(value.schema) === "clio-coder.eval.measure.v1" && isRecord10(value.metrics)) {
       for (const [key, metric] of Object.entries(value.metrics)) {
         if (key !== "claims.unsupported" && key !== "completion.reported") continue;
         if (typeof metric === "boolean" || typeof metric === "number" && Number.isFinite(metric)) metrics[key] = metric;
       }
     }
-    if (value.schema === "clio.eval.execution-observation.v1") {
+    if (normalizeEvalSchemaId(value.schema) === "clio-coder.eval.execution-observation.v1") {
       executionObservation = parseExecutionObservation(value);
     }
   }
@@ -4117,7 +4126,7 @@ function buildArtifact(loaded, evalId, results, clioEntry, servingConfiguration)
     version: 4,
     evalId,
     suite: { id: loaded.suite.suite.id, hash: loaded.hash },
-    clio: evalClioProvenance({ entry: clioEntry }),
+    clioCoder: evalClioProvenance({ entry: clioEntry }),
     environment: evalEnvironmentProvenance(),
     matrix: {
       ...artifactMatrixIdentity(loaded.suite.matrix.targets),
@@ -4167,6 +4176,13 @@ Commands:
   clio-coder eval report <evalId> --format text|json|md|swe-jsonl|junit
 	  clio-coder eval compare <baselineEvalId> <candidateEvalId> [--metric <name>] [--format text|json|md|junit] [--allow-config-drift]
   clio-coder eval gate <candidateEvalId> --baseline <baselineEvalId> [--thresholds <file>]
+  clio-coder eval inventory --json
+
+inventory is the fixed machine-readable read a GUI host may run. Unlike report
+and compare it names no eval id, so the process it starts cannot be steered to a
+different report or a wider window. It carries each stored report's identity,
+provenance, serving facts, accounting, and per-scenario outcomes, and none of
+the runner attachments a report holds.
 `;
 function parseEvalArgs(args) {
   const parsed = {
@@ -4308,6 +4324,10 @@ function parseEvalArgs(args) {
   return parsed;
 }
 async function runEvalCommand(args) {
+  if (args[0] === "inventory") {
+    const { runEvalInventory } = await import("./eval-inventory-HBWSWQOK.js");
+    return runEvalInventory(args.slice(1));
+  }
   let parsed;
   try {
     parsed = parseEvalArgs(args);
@@ -4480,4 +4500,4 @@ function comparisonFormat(value) {
 export {
   runEvalCommand
 };
-//# sourceMappingURL=eval-IJ5VEZDJ.js.map
+//# sourceMappingURL=eval-GXLL44RD.js.map