@bastani/atomic 0.8.26-alpha.5 → 0.8.26-alpha.6

package/dist/builtin/subagents/src/runs/shared/model-fallback.ts

@@ -62,7 +62,7 @@ export function currentModelFullId(model: { provider: string; id: string } | und
 	return `${String(model.provider)}/${model.id}`;
 }
 
-const RETRYABLE_MODEL_FAILURE_PATTERNS = [
+const RETRYABLE_MODEL_FAILURE_PATTERNS: readonly RegExp[] = [
 	/rate\s*limit/i,
 	/too many requests/i,
 	/\b429\b/,
@@ -71,6 +71,7 @@ const RETRYABLE_MODEL_FAILURE_PATTERNS = [
 	/credit/i,
 	/auth(?:entication)?/i,
 	/unauthori[sz]ed/i,
+	/\b40[13]\b/,
 	/forbidden/i,
 	/api key/i,
 	/token expired/i,
@@ -90,14 +91,422 @@ const RETRYABLE_MODEL_FAILURE_PATTERNS = [
 	/upstream/i,
 	/timed? out/i,
 	/timeout/i,
-	/\b502\b/,
-	/\b503\b/,
-	/\b504\b/,
+	/\b50[0-4]\b/,
 ];
 
-export function isRetryableModelFailure(error: string | undefined): boolean {
-	if (!error) return false;
-	return RETRYABLE_MODEL_FAILURE_PATTERNS.some((pattern) => pattern.test(error));
+const NON_RETRYABLE_FAILURE_PATTERNS: readonly RegExp[] = [
+	/command failed/i,
+	/tests? failed/i,
+	/shell/i,
+	/missing file/i,
+	/no such file/i,
+	/completion guard/i,
+	/cancel/i,
+	/abort/i,
+	/interrupted/i,
+];
+
+const CANCELLED_FAILURE_PATTERNS: readonly RegExp[] = [
+	/cancel/i,
+	/abort/i,
+	/interrupted/i,
+];
+
+export type ModelFallbackFailureKind =
+	| "auth_on_candidate_provider"
+	| "rate_limit"
+	| "provider_unavailable"
+	| "network_timeout"
+	| "model_unavailable"
+	| "cancelled"
+	| "task_failure"
+	| "unknown";
+
+export type ModelFallbackFailureSource =
+	| "assistant_message"
+	| "diagnostic"
+	| "throw"
+	| "structured"
+	| "string_fallback";
+
+export interface ModelFallbackFailureSignal {
+	readonly kind: ModelFallbackFailureKind;
+	readonly message: string;
+	readonly source: ModelFallbackFailureSource;
+	readonly stopReason?: string;
+	readonly status?: number;
+	readonly code?: string | number;
+	readonly name?: string;
+}
+
+const FALLBACKABLE_FAILURE_KINDS: ReadonlySet<ModelFallbackFailureKind> = new Set([
+	"auth_on_candidate_provider",
+	"rate_limit",
+	"provider_unavailable",
+	"network_timeout",
+	"model_unavailable",
+]);
+
+function asRecord(value: unknown): Record<string, unknown> | undefined {
+	return value !== null && typeof value === "object" ? value as Record<string, unknown> : undefined;
+}
+
+function field(value: unknown, key: string): unknown {
+	return asRecord(value)?.[key];
+}
+
+function stringField(value: unknown, key: string): string | undefined {
+	const raw = field(value, key);
+	return typeof raw === "string" && raw.trim().length > 0 ? raw : undefined;
+}
+
+function errorName(value: unknown): string | undefined {
+	return value instanceof Error ? value.name : stringField(value, "name");
+}
+
+function directMessageFrom(value: unknown): string | undefined {
+	return stringField(value, "errorMessage")
+		?? stringField(value, "message")
+		?? stringField(value, "statusText");
+}
+
+function integerFrom(value: unknown): number | undefined {
+	if (typeof value === "number" && Number.isInteger(value)) return value;
+	if (typeof value !== "string" || value.trim().length === 0) return undefined;
+	const parsed = Number(value.trim());
+	return Number.isInteger(parsed) ? parsed : undefined;
+}
+
+function statusFrom(value: unknown): number | undefined {
+	return integerFrom(field(value, "status"))
+		?? integerFrom(field(value, "statusCode"))
+		?? integerFrom(field(value, "httpStatus"));
+}
+
+function codeFrom(value: unknown): string | number | undefined {
+	const rawCode = field(value, "code");
+	return typeof rawCode === "string" || typeof rawCode === "number" ? rawCode : undefined;
+}
+
+function stopReasonFrom(value: unknown): string | undefined {
+	return stringField(value, "stopReason");
+}
+
+function finishReasonFrom(value: unknown): string | undefined {
+	return stringField(value, "finish_reason") ?? stringField(value, "finishReason");
+}
+
+function causeOf(value: unknown): unknown {
+	return value instanceof Error ? value.cause : field(value, "cause");
+}
+
+function diagnosticErrors(value: unknown): readonly unknown[] {
+	const diagnostics = field(value, "diagnostics");
+	if (!Array.isArray(diagnostics)) return [];
+	const errors: unknown[] = [];
+	for (const diagnostic of diagnostics) {
+		const diagnosticError = field(diagnostic, "error");
+		errors.push(diagnosticError ?? diagnostic);
+	}
+	return errors;
+}
+
+function normalizeCode(value: string | number | undefined): string | undefined {
+	if (value === undefined) return undefined;
+	const normalized = String(value)
+		.trim()
+		.toLowerCase()
+		.replace(/[^a-z0-9]+/g, "_")
+		.replace(/^_+|_+$/g, "");
+	return normalized.length > 0 ? normalized : undefined;
+}
+
+function kindFromStatus(status: number | undefined): ModelFallbackFailureKind | undefined {
+	switch (status) {
+		case 401:
+		case 403:
+			return "auth_on_candidate_provider";
+		case 408:
+			return "network_timeout";
+		case 404:
+			return "model_unavailable";
+		case 429:
+			return "rate_limit";
+		default:
+			if (status !== undefined && status >= 500 && status <= 599) return "provider_unavailable";
+			return undefined;
+	}
+}
+
+function refusalKindFromCode(code: string | number | undefined): ModelFallbackFailureKind | undefined {
+	const normalizedCode = normalizeCode(code);
+	if (normalizedCode === undefined) return undefined;
+	if (normalizedCode.includes("content_filter") || normalizedCode.includes("contentfilter")) return "task_failure";
+	if (normalizedCode.includes("safety") || normalizedCode.includes("policy")) return "task_failure";
+	switch (normalizedCode) {
+		case "blocked":
+		case "blocked_by_provider":
+		case "blocked_by_safety":
+		case "blocked_by_policy":
+		case "provider_refusal":
+		case "refusal":
+		case "tool_refusal":
+		case "tool_call_refusal":
+		case "tool_use_refusal":
+			return "task_failure";
+		default:
+			return undefined;
+	}
+}
+
+function kindFromCode(code: string | number | undefined): ModelFallbackFailureKind | undefined {
+	const normalizedCode = normalizeCode(code);
+	if (normalizedCode === undefined) return undefined;
+	const refusalKind = refusalKindFromCode(code);
+	if (refusalKind !== undefined) return refusalKind;
+	const httpStatusKind = kindFromStatus(integerFrom(code));
+	if (httpStatusKind !== undefined) return httpStatusKind;
+
+	switch (normalizedCode) {
+		case "auth":
+		case "auth_required":
+		case "authentication_required":
+		case "unauthorized":
+		case "forbidden":
+		case "invalid_api_key":
+		case "missing_api_key":
+		case "invalid_key":
+			return "auth_on_candidate_provider";
+		case "etimedout":
+		case "econnreset":
+		case "econnrefused":
+		case "enotfound":
+		case "eai_again":
+		case "fetch_failed":
+		case "network_error":
+		case "timeout":
+		case "timeout_error":
+		case "und_err_connect_timeout":
+			return "network_timeout";
+		case "rate_limit":
+		case "rate_limit_exceeded":
+		case "too_many_requests":
+		case "quota_exceeded":
+			return "rate_limit";
+		case "aborterror":
+		case "aborted":
+		case "cancelled":
+		case "canceled":
+			return "cancelled";
+		case "model_not_found":
+		case "model_unavailable":
+		case "model_disabled":
+		case "unknown_model":
+			return "model_unavailable";
+		case "provider_error":
+		case "api_error":
+		case "service_unavailable":
+		case "temporarily_unavailable":
+		case "overloaded":
+			return "provider_unavailable";
+		default:
+			return undefined;
+	}
+}
+
+const PROVIDER_REFUSAL_FAILURE_PATTERNS: readonly RegExp[] = [
+	/\bfinish[_\s-]?reason\b[^\n]*\bcontent[_\s-]?filter\b/i,
+	/\bcontent[_\s-]?filter(?:ed|ing)?\b/i,
+	/\b(?:safety|policy)\b[^\n]*\b(?:refus(?:e|al|ed|es|ing)?|block(?:ed|ing)?|filter(?:ed|ing)?|violat(?:e|ion|ed|ing)?|disallow(?:ed|ing)?|reject(?:ed|ion|ing)?)\b/i,
+	/\b(?:refus(?:e|al|ed|es|ing)?|block(?:ed|ing)?|filter(?:ed|ing)?|violat(?:e|ion|ed|ing)?|disallow(?:ed|ing)?|reject(?:ed|ion|ing)?)\b[^\n]*\b(?:safety|policy)\b/i,
+	/\btool[_\s-]?(?:call|use)?[_\s-]?refus(?:e|al|ed|es|ing)?\b/i,
+	/\btool(?:\s+call|\s+use)?\b[^\n]*\brefus(?:e|al|ed|es|ing)?\b/i,
+	/\brefus(?:e|al|ed|es|ing)?\b[^\n]*\btool(?:\s+call|\s+use)?\b/i,
+	/\bprovider[_\s-]?refus(?:e|al|ed|es|ing)?\b/i,
+	/\bprovider\b[^\n]*\brefus(?:e|al|ed|es|ing)?\b[^\n]*\b(?:prompt|request|content|policy|safety)\b/i,
+];
+
+function refusalKindFromMessage(message: string): ModelFallbackFailureKind | undefined {
+	if (CANCELLED_FAILURE_PATTERNS.some((pattern) => pattern.test(message))) return "cancelled";
+	if (NON_RETRYABLE_FAILURE_PATTERNS.some((pattern) => pattern.test(message))) return "task_failure";
+	if (PROVIDER_REFUSAL_FAILURE_PATTERNS.some((pattern) => pattern.test(message))) return "task_failure";
+	return undefined;
+}
+
+function fallbackKindFromMessage(message: string, name: string | undefined): ModelFallbackFailureKind | undefined {
+	const refusalKind = refusalKindFromMessage(message);
+	if (refusalKind !== undefined) return refusalKind;
+	const nameKind = kindFromCode(name);
+	if (nameKind !== undefined) return nameKind;
+	if (!RETRYABLE_MODEL_FAILURE_PATTERNS.some((pattern) => pattern.test(message))) return undefined;
+	if (/rate\s*limit|too many requests|\b429\b|quota|billing|credit/i.test(message)) return "rate_limit";
+	if (/auth|unauthori[sz]ed|\b40[13]\b|api key|token expired|forbidden|invalid key/i.test(message)) return "auth_on_candidate_provider";
+	if (/model.*(?:unavailable|disabled|not found|unknown)|(?:unavailable|disabled|not found|unknown).*model/i.test(message)) return "model_unavailable";
+	if (/network|fetch failed|socket|connection refused|timeout|timed? out/i.test(message)) return "network_timeout";
+	return "provider_unavailable";
+}
+
+function signalSource(value: unknown, fallback: ModelFallbackFailureSource | undefined): ModelFallbackFailureSource {
+	if (fallback !== undefined) return fallback;
+	if (stopReasonFrom(value) !== undefined || diagnosticErrors(value).length > 0) return "assistant_message";
+	if (value instanceof Error) return "throw";
+	return "structured";
+}
+
+function makeSignal(
+	kind: ModelFallbackFailureKind,
+	value: unknown,
+	source: ModelFallbackFailureSource | undefined,
+): ModelFallbackFailureSignal {
+	const status = statusFrom(value);
+	const code = codeFrom(value);
+	const name = errorName(value);
+	const stopReason = stopReasonFrom(value);
+	return {
+		kind,
+		message: modelFailureMessage(value),
+		source: signalSource(value, source),
+		...(stopReason !== undefined ? { stopReason } : {}),
+		...(status !== undefined ? { status } : {}),
+		...(code !== undefined ? { code } : {}),
+		...(name !== undefined ? { name } : {}),
+	};
+}
+
+function fallbackSignalFromMessage(
+	value: unknown,
+	source: ModelFallbackFailureSource | undefined,
+): ModelFallbackFailureSignal | undefined {
+	const message = modelFailureMessage(value);
+	if (!message.trim()) return undefined;
+	const kind = fallbackKindFromMessage(message, errorName(value));
+	return kind === undefined ? undefined : makeSignal(kind, value, source);
+}
+
+function classifyAssistantRefusalSignal(
+	value: unknown,
+	source: ModelFallbackFailureSource | undefined,
+): ModelFallbackFailureSignal | undefined {
+	const codeRefusalKind = refusalKindFromCode(codeFrom(value))
+		?? refusalKindFromCode(errorName(value))
+		?? refusalKindFromCode(finishReasonFrom(value));
+	if (codeRefusalKind !== undefined) return makeSignal(codeRefusalKind, value, source);
+
+	const messageRefusalKind = refusalKindFromMessage(directMessageFrom(value) ?? "");
+	return messageRefusalKind === undefined ? undefined : makeSignal(messageRefusalKind, value, source);
+}
+
+function isRefusalSignal(signal: ModelFallbackFailureSignal): boolean {
+	return signal.kind === "cancelled" || signal.kind === "task_failure";
+}
+
+function structuredSignal(
+	value: unknown,
+	seen: Set<unknown>,
+	source?: ModelFallbackFailureSource,
+): ModelFallbackFailureSignal | undefined {
+	if (value === undefined || value === null || seen.has(value)) return undefined;
+	if (typeof value === "object") seen.add(value);
+
+	const stopReason = stopReasonFrom(value)?.toLowerCase();
+	if (stopReason === "aborted") return makeSignal("cancelled", value, source);
+
+	const directRefusalSignal = classifyAssistantRefusalSignal(value, source);
+	if (directRefusalSignal !== undefined) return directRefusalSignal;
+
+	const codeKind = kindFromCode(codeFrom(value));
+	const nameKind = kindFromCode(errorName(value));
+	if (codeKind === "cancelled" || nameKind === "cancelled") return makeSignal("cancelled", value, source);
+
+	let firstNestedFallbackSignal: ModelFallbackFailureSignal | undefined;
+	const nestedSeen = new Set(seen);
+	for (const diagnosticError of diagnosticErrors(value)) {
+		const diagnosticSignal = structuredSignal(diagnosticError, nestedSeen, "diagnostic")
+			?? fallbackSignalFromMessage(diagnosticError, "diagnostic");
+		if (diagnosticSignal === undefined) continue;
+		if (isRefusalSignal(diagnosticSignal)) return diagnosticSignal;
+		firstNestedFallbackSignal ??= diagnosticSignal;
+	}
+
+	const cause = causeOf(value);
+	const causeSignal = structuredSignal(cause, nestedSeen, source)
+		?? fallbackSignalFromMessage(cause, source);
+	if (causeSignal !== undefined) {
+		if (isRefusalSignal(causeSignal)) return causeSignal;
+		firstNestedFallbackSignal ??= causeSignal;
+	}
+
+	const statusKind = kindFromStatus(statusFrom(value));
+	if (statusKind !== undefined) return makeSignal(statusKind, value, source);
+	if (codeKind !== undefined) return makeSignal(codeKind, value, source);
+	if (nameKind !== undefined) return makeSignal(nameKind, value, source);
+
+	if (firstNestedFallbackSignal !== undefined) return firstNestedFallbackSignal;
+
+	if (stopReason === "error") return makeSignal("provider_unavailable", value, source);
+
+	return undefined;
+}
+
+function messageFromUnknown(value: unknown, seen: Set<unknown>): string | undefined {
+	if (value === undefined || value === null || seen.has(value)) return undefined;
+	if (typeof value === "string") return value.trim().length > 0 ? value : undefined;
+	if (typeof value === "number" || typeof value === "boolean" || typeof value === "bigint") return String(value);
+	if (typeof value === "symbol" || typeof value === "function") return undefined;
+	seen.add(value);
+
+	if (value instanceof Error && value.message.trim().length > 0) return value.message;
+	const directMessage = directMessageFrom(value);
+	if (directMessage !== undefined) return directMessage;
+
+	for (const diagnosticError of diagnosticErrors(value)) {
+		const diagnosticMessage = messageFromUnknown(diagnosticError, seen);
+		if (diagnosticMessage !== undefined) return diagnosticMessage;
+	}
+
+	const causeMessage = messageFromUnknown(causeOf(value), seen);
+	if (causeMessage !== undefined) return causeMessage;
+
+	const stopReason = stopReasonFrom(value);
+	if (stopReason !== undefined) return `Assistant message ended with stopReason:${stopReason}`;
+	const finishReason = finishReasonFrom(value);
+	if (finishReason !== undefined) return `Model request finished with finish_reason:${finishReason}`;
+	const status = statusFrom(value);
+	if (status !== undefined) return `Model request failed with status ${status}`;
+	const code = codeFrom(value);
+	if (code !== undefined) return `Model request failed with code ${String(code)}`;
+
+	return undefined;
+}
+
+export function modelFailureMessage(error: unknown): string {
+	const structuredMessage = messageFromUnknown(error, new Set());
+	if (structuredMessage !== undefined) return structuredMessage;
+	const rendered = String(error);
+	return rendered === "[object Object]" ? "Model request failed" : rendered;
+}
+
+export function normalizeModelFailureSignal(error: unknown): ModelFallbackFailureSignal {
+	const structured = structuredSignal(error, new Set());
+	if (structured !== undefined) return structured;
+
+	const message = modelFailureMessage(error);
+	const name = errorName(error);
+	const fallbackKind = message.trim().length > 0
+		? fallbackKindFromMessage(message, name)
+		: undefined;
+	return {
+		kind: fallbackKind ?? "unknown",
+		message,
+		source: "string_fallback",
+		...(name !== undefined ? { name } : {}),
+	};
+}
+
+export function isRetryableModelFailure(error: unknown): boolean {
+	if (error === undefined) return false;
+	const signal = normalizeModelFailureSignal(error);
+	return FALLBACKABLE_FAILURE_KINDS.has(signal.kind);
 }
 
 export function formatModelAttemptNote(attempt: ModelAttemptSummary, nextModel?: string): string {

package/dist/builtin/web-access/CHANGELOG.md

@@ -4,6 +4,12 @@ All notable changes to this project will be documented in this file.
 
 ## [Unreleased]
 
+## [0.8.26-alpha.6] - 2026-06-06
+
+### Changed
+
+- Bumped package version for the Atomic 0.8.26-alpha.6 prerelease.
+
 ## [0.8.26-alpha.5] - 2026-06-06
 
 ### Changed

package/dist/builtin/web-access/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bastani/web-access",
-  "version": "0.8.26-alpha.5",
+  "version": "0.8.26-alpha.6",
   "private": true,
   "description": "Atomic extension for web search, URL fetching, GitHub repo cloning, PDF/video extraction. Fork of: https://github.com/nicobailon/pi-web-access",
   "contributors": [

package/dist/builtin/workflows/CHANGELOG.md

@@ -6,6 +6,12 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 
 ## [Unreleased]
 
+## [0.8.26-alpha.6] - 2026-06-06
+
+### Changed
+
+- Bumped package version for the Atomic 0.8.26-alpha.6 prerelease.
+
 ## [0.8.26-alpha.5] - 2026-06-06
 
 ### Fixed

package/dist/builtin/workflows/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bastani/workflows",
-  "version": "0.8.26-alpha.5",
+  "version": "0.8.26-alpha.6",
   "private": true,
   "description": "Atomic extension for multi-stage workflow authoring and execution.",
   "contributors": [

package/dist/builtin/workflows/src/extension/index.ts

@@ -1795,7 +1795,11 @@ export function makeExecuteWorkflowTool(
         const isPaused =
           run?.status === "paused" ||
           (run?.stages.some((s) => s.status === "paused") ?? false);
-        if (!isPaused && run?.status === "failed" && run.endedAt !== undefined && run.resumable !== false) {
+        const isResumableContinuation = run !== undefined && !isPaused && (
+          (run.status === "failed" && run.endedAt !== undefined && run.resumable !== false) ||
+          (run.endedAt === undefined && run.resumable === true && run.failureRecoverability === "recoverable")
+        );
+        if (isResumableContinuation) {
           const continuation = activeRuntime.resumeFailedRun(stageRunId, stage.stageId, { policy });
           return {
             action: "resume",
@@ -3133,7 +3137,11 @@ function factory(pi: ExtensionAPI): void {
       const isPaused =
         run?.status === "paused" ||
         (run?.stages.some((s) => s.status === "paused") ?? false);
-      if (!isPaused && run?.status === "failed" && run.endedAt !== undefined && run.resumable !== false) {
+      const isResumableContinuation = run !== undefined && !isPaused && (
+        (run.status === "failed" && run.endedAt !== undefined && run.resumable !== false) ||
+        (run.endedAt === undefined && run.resumable === true && run.failureRecoverability === "recoverable")
+      );
+      if (isResumableContinuation) {
         const continuation = runtimeForContext(ctx).resumeFailedRun(stageRunId, stageId, { policy });
         if (continuation.ok) {
           print(continuation.message);

package/dist/builtin/workflows/src/extension/runtime.ts

@@ -44,6 +44,7 @@ import {
 import { validateWorkflowModels } from "../runs/shared/model-fallback.js";
 import { runDetached } from "../runs/background/runner.js";
 import type { JobTracker } from "../runs/background/job-tracker.js";
+import { appendRunEnd } from "../shared/persistence-session-entries.js";
 import { classifyWorkflowFailure } from "../shared/workflow-failures.js";
 
 // ---------------------------------------------------------------------------
@@ -389,13 +390,39 @@ export function createExtensionRuntime(opts: ExtensionRuntimeOpts = {}): Extensi
     return { ok: true, stageId: failedStageId };
   }
 
+  function finalizeResumedActiveBlockedSourceRun(source: RunSnapshot, continuationRunId: string): void {
+    const errorMessage = source.error ?? source.failureMessage ?? `workflow resumed in new run ${continuationRunId}`;
+    const metadata = {
+      ...(source.failureKind !== undefined ? { failureKind: source.failureKind } : {}),
+      ...(source.failureCode !== undefined ? { failureCode: source.failureCode } : {}),
+      failureRecoverability: "non_recoverable",
+      failureDisposition: "terminal_killed",
+      ...(source.failureMessage !== undefined ? { failureMessage: source.failureMessage } : {}),
+      ...(source.failedStageId !== undefined ? { failedStageId: source.failedStageId } : {}),
+      resumable: false,
+      ...(source.retryAfterMs !== undefined ? { retryAfterMs: source.retryAfterMs } : {}),
+    } as const;
+    const recorded = activeStore.recordRunEnd(source.id, "killed", undefined, errorMessage, metadata);
+    if (recorded && persistence !== undefined) {
+      appendRunEnd(persistence, {
+        runId: source.id,
+        status: "killed",
+        error: errorMessage,
+        ...metadata,
+        ts: Date.now(),
+      });
+    }
+  }
+
   function resumeFailedRun(sourceRunId: string, stageId?: string, options?: RuntimeDispatchOptions): ResumeFailedRunResult {
     const source = activeStore.runs().find((run) => run.id === sourceRunId);
     if (source === undefined) {
       return { ok: false, reason: "run_not_found", message: `run not found: ${sourceRunId}` };
     }
-    if (source.status !== "failed" || source.endedAt === undefined || source.resumable === false) {
-      return { ok: false, reason: "not_resumable", message: `run ${sourceRunId} is not a failed resumable workflow run` };
+    const isTerminalFailedResumable = source.status === "failed" && source.endedAt !== undefined && source.resumable !== false;
+    const isActiveBlockedResumable = source.endedAt === undefined && source.resumable === true && source.failureRecoverability === "recoverable";
+    if (!isTerminalFailedResumable && !isActiveBlockedResumable) {
+      return { ok: false, reason: "not_resumable", message: `run ${sourceRunId} is not a resumable workflow run` };
     }
     const def = registry.get(source.name);
     if (def === undefined) {
@@ -415,12 +442,17 @@ export function createExtensionRuntime(opts: ExtensionRuntimeOpts = {}): Extensi
       ...runOptions({ workflow: def.name, inputs: sourceInputs }, options?.policy),
       continuation: { source, resumeFromStageId: resolvedStage.stageId },
     });
+    if (isActiveBlockedResumable) {
+      finalizeResumedActiveBlockedSourceRun(source, accepted.runId);
+    }
     return {
       ok: true,
       runId: accepted.runId,
       sourceRunId: source.id,
       resumeFromStageId: resolvedStage.stageId,
-      message: `Resuming failed workflow "${def.name}" from run ${source.id.slice(0, 8)} at stage ${resolvedStage.stageId.slice(0, 8)} (new run ${accepted.runId}).`,
+      message: isActiveBlockedResumable
+        ? `Resuming blocked workflow "${def.name}" from run ${source.id.slice(0, 8)} at stage ${resolvedStage.stageId.slice(0, 8)} (new run ${accepted.runId}).`
+        : `Resuming failed workflow "${def.name}" from run ${source.id.slice(0, 8)} at stage ${resolvedStage.stageId.slice(0, 8)} (new run ${accepted.runId}).`,
     };
   }
 

package/dist/builtin/workflows/src/runs/background/status.ts

@@ -82,6 +82,14 @@ export interface RunDetail {
   readonly stages: readonly RunSnapshot["stages"][number][];
   readonly result?: WorkflowOutputValues;
   readonly error?: string;
+  readonly failureKind?: RunSnapshot["failureKind"];
+  readonly failureCode?: RunSnapshot["failureCode"];
+  readonly failureRecoverability?: RunSnapshot["failureRecoverability"];
+  readonly failureDisposition?: RunSnapshot["failureDisposition"];
+  readonly failedStageId?: string;
+  readonly resumable?: boolean;
+  readonly retryAfterMs?: number;
+  readonly blockedAt?: number;
 }
 
 export type InspectRunResult =
@@ -146,11 +154,26 @@ export function killRun(
   const previousStatus = run.status;
 
   // Abort active executor (no-op if not registered)
-  opts?.cancellation?.abort(runId, "workflow killed");
-
-  const recorded = activeStore.recordRunEnd(runId, "killed", undefined, "workflow killed");
+  const errorMessage = "workflow killed";
+  opts?.cancellation?.abort(runId, errorMessage);
+
+  const metadata = {
+    failureKind: "cancelled",
+    failureCode: "cancelled",
+    failureRecoverability: "non_recoverable",
+    failureDisposition: "terminal_killed",
+    failureMessage: errorMessage,
+    resumable: false,
+  } as const;
+  const recorded = activeStore.recordRunEnd(runId, "killed", undefined, errorMessage, metadata);
   if (recorded && opts?.persistence) {
-    appendRunEnd(opts.persistence, { runId, status: "killed", ts: Date.now() });
+    appendRunEnd(opts.persistence, {
+      runId,
+      status: "killed",
+      error: errorMessage,
+      ...metadata,
+      ts: Date.now(),
+    });
   }
 
   return { ok: true, runId, previousStatus };
@@ -247,14 +270,29 @@ export function resumeRun(
   // Return a deep copy of the snapshot for safe consumption
   const snapshot = structuredClone(run);
   const resumedCopy = structuredClone(resumed);
-  if (run.status === "failed" && run.endedAt !== undefined && run.resumable === false) {
+  if (run.status === "killed" || run.resumable === false) {
     return {
       ok: true,
       runId,
       snapshot,
       resumed: resumedCopy,
       mode: "not_resumable",
-      message: "This failed workflow is not resumable; inspect the snapshot and rerun the workflow when ready.",
+      message: "This workflow is not resumable; inspect the snapshot and start a new workflow run when ready.",
+    };
+  }
+  if (
+    run.endedAt === undefined &&
+    run.resumable === true &&
+    run.failureRecoverability === "recoverable" &&
+    run.failedStageId !== undefined
+  ) {
+    return {
+      ok: true,
+      runId,
+      snapshot,
+      resumed: resumedCopy,
+      mode: resumedCopy.length > 0 ? "paused" : "snapshot",
+      message: `Workflow is blocked on a recoverable ${run.failureCode ?? run.failureKind ?? "workflow"} failure at stage ${run.failedStageId}; retry/resume after the issue clears.`,
     };
   }
   return {
@@ -441,6 +479,14 @@ export function inspectRun(
     stages: expandedStages.map((stage) => structuredClone(stage)),
     result: copy.result,
     error: copy.error,
+    failureKind: copy.failureKind,
+    failureCode: copy.failureCode,
+    failureRecoverability: copy.failureRecoverability,
+    failureDisposition: copy.failureDisposition,
+    failedStageId: copy.failedStageId,
+    resumable: copy.resumable,
+    retryAfterMs: copy.retryAfterMs,
+    blockedAt: copy.blockedAt,
   };
 
   return { ok: true, runId: copy.id, detail };