dsh-llm-verifier 0.6.7 → 0.6.9

package/lib/{auto-DkUMv8FV.js → auto-C9YG0E22.js}

@@ -361,7 +361,7 @@ function blockText$1(blocks) {
 * @param blocks - content blocks of an `assistant/message` event.
 * @returns The joined text, trimmed.
 */
-function narrativeText(blocks) {
+function narrativeText$1(blocks) {
 	const parts = [];
 	const visit = (items) => {
 		for (const block of items) if (block.type === "text") parts.push(block.text);
@@ -433,7 +433,7 @@ function buildEvidenceIndex(events) {
 			}
 		} else if (event.type === "assistant/message") {
 			const blocks = event.data?.message?.content ?? event.data?.content;
-			const text = narrativeText(Array.isArray(blocks) ? blocks : []);
+			const text = narrativeText$1(Array.isArray(blocks) ? blocks : []);
 			if (text) narration = {
 				seq: event.seq,
 				text
@@ -648,7 +648,7 @@ const VERIFIER_EVIDENCE_TOOLS = /* @__PURE__ */ new Set([
 	"verifier_current_session"
 ]);
 /** Tools that return either a child's report (evidence) or a bare start acknowledgement (not evidence). */
-const SUBAGENT_TOOLS = /* @__PURE__ */ new Set(["subagent", "subagent_fork"]);
+const SUBAGENT_TOOLS$1 = /* @__PURE__ */ new Set(["subagent", "subagent_fork"]);
 /**
 * Start acknowledgement of a background child.
 *
@@ -672,7 +672,7 @@ function isEvidenceOutput(name, text) {
 	if (BOOKKEEPING_TOOLS.has(name) || VERIFIER_EVIDENCE_TOOLS.has(name)) return false;
 	const body = text.trim();
 	if (!body) return false;
-	return !(SUBAGENT_TOOLS.has(name) && CHILD_START_ACKNOWLEDGEMENT.test(body));
+	return !(SUBAGENT_TOOLS$1.has(name) && CHILD_START_ACKNOWLEDGEMENT.test(body));
 }
 /** Signatures of an output that reports a PASSED verification result. */
 const VERIFICATION_PASS_SIGNATURES = [
@@ -1776,7 +1776,8 @@ const PASSIVE_TOOLS = /* @__PURE__ */ new Set([
 	"list_agents",
 	"get_goal",
 	"skill",
-	"mcp__codegraph__codegraph_explore"
+	"mcp__codegraph__codegraph_explore",
+	"ask_user_question"
 ]);
 const VERIFIER_TOOLS = /* @__PURE__ */ new Set([
 	"verifier_compare",
@@ -1883,6 +1884,236 @@ function isSuccessfulCodeDispatch(data) {
 	if (Array.isArray(data.content) && data.content.some((b) => b.isError === true)) return false;
 	return true;
 }
+const SUBAGENT_TOOLS = /* @__PURE__ */ new Set(["subagent", "subagent_fork"]);
+const CONTINUABLE_SUBAGENT_START = /^\s*started subagent\s+([^\s`]+)/imu;
+const BACKGROUND_SUBAGENT_JOB_START = /^\s*started background subagent job\s+([^\s`]+)/imu;
+const GENERAL_SUBAGENT_START = /^\s*started (?:background )?subagent\b/imu;
+/**
+* Whether background subagents started during the current task remain in flight.
+*
+* A background subagent returns immediately with a start receipt (`started subagent <id>`
+* or `started background subagent job <id>`) and settles later via a runtime-injected notice
+* (`source.kind === 'subagent-settled'` or a job settlement notice). Performing session
+* acceptance while subagents are in flight will always fail and steer prematurely because the
+* delegated work has not reported back yet.
+* @param events - session events.
+* @param taskStartSeq - sequence number of the current direct user task statement.
+* @returns True when at least one background subagent remains unsettled.
+*/
+function hasPendingSubagents(events, taskStartSeq) {
+	const relevant = events.filter((event) => event.seq >= taskStartSeq);
+	const pendingContinuable = /* @__PURE__ */ new Map();
+	const pendingJobs = /* @__PURE__ */ new Map();
+	let anonymousPendingCount = 0;
+	const calls = /* @__PURE__ */ new Map();
+	for (const event of relevant) if (event.type === "tool/call") calls.set(String(event.data.callId), event);
+	for (const event of relevant) {
+		if (event.type === "tool/result") {
+			const call = calls.get(String(event.data.message.source.callId));
+			if (call && SUBAGENT_TOOLS.has(call.data.name) && event.data.error === void 0) {
+				if (event.data.message.content.every((b) => b.isError !== true)) {
+					const text = blockText(event.data.message.content);
+					const contMatch = text.match(CONTINUABLE_SUBAGENT_START);
+					const jobMatch = text.match(BACKGROUND_SUBAGENT_JOB_START);
+					if (contMatch) pendingContinuable.set(contMatch[1], event.seq);
+					else if (jobMatch) pendingJobs.set(jobMatch[1], event.seq);
+					else if (GENERAL_SUBAGENT_START.test(text)) anonymousPendingCount += 1;
+				}
+			}
+		} else if (CODE_DISPATCH_TYPES.has(event.type)) {
+			const data = event.data;
+			if (data && SUBAGENT_TOOLS.has(data.name) && isSuccessfulCodeDispatch(data)) {
+				const text = blockText(data.content);
+				const contMatch = text.match(CONTINUABLE_SUBAGENT_START);
+				const jobMatch = text.match(BACKGROUND_SUBAGENT_JOB_START);
+				if (contMatch) pendingContinuable.set(contMatch[1], event.seq);
+				else if (jobMatch) pendingJobs.set(jobMatch[1], event.seq);
+				else if (GENERAL_SUBAGENT_START.test(text)) anonymousPendingCount += 1;
+			}
+		}
+		if (event.type === "user/message") {
+			const source = event.data?.source;
+			const text = blockText(event.data?.content);
+			if (source?.kind === "subagent-settled") {
+				const senderId = typeof source.senderSessionId === "string" ? source.senderSessionId : void 0;
+				if (senderId && pendingContinuable.has(senderId)) pendingContinuable.delete(senderId);
+				else {
+					let found = false;
+					for (const [id] of pendingContinuable) if (text.includes(id)) {
+						pendingContinuable.delete(id);
+						found = true;
+						break;
+					}
+					if (!found) {
+						if (pendingContinuable.size > 0) {
+							const oldest = pendingContinuable.keys().next().value;
+							if (oldest !== void 0) pendingContinuable.delete(oldest);
+						} else if (anonymousPendingCount > 0) anonymousPendingCount -= 1;
+					}
+				}
+			} else if (source?.plugin === "tool-jobs" || /background job\b.*finished/iu.test(text)) {
+				for (const [jobId] of pendingJobs) if (text.includes(jobId)) {
+					pendingJobs.delete(jobId);
+					break;
+				}
+			}
+		}
+		if (event.type === "tool/call") {
+			const name = event.data.name;
+			const argsText = event.data.arguments;
+			let parsedArgs = {};
+			if (typeof argsText === "string") try {
+				parsedArgs = JSON.parse(argsText);
+			} catch {}
+			else if (typeof argsText === "object" && argsText !== null) parsedArgs = argsText;
+			if (name === "interrupt_agent") {
+				const target = String(parsedArgs.target ?? parsedArgs.agent_id ?? "");
+				if (target && pendingContinuable.has(target)) pendingContinuable.delete(target);
+			} else if (name === "job_kill") {
+				const jobId = String(parsedArgs.job_id ?? "");
+				if (jobId && pendingJobs.has(jobId)) pendingJobs.delete(jobId);
+			}
+		} else if (CODE_DISPATCH_TYPES.has(event.type)) {
+			const data = event.data;
+			if (data) {
+				const name = data.name;
+				let parsedArgs = {};
+				if (typeof data.arguments === "string") try {
+					parsedArgs = JSON.parse(data.arguments);
+				} catch {}
+				else if (typeof data.arguments === "object" && data.arguments !== null) parsedArgs = data.arguments;
+				if (name === "interrupt_agent") {
+					const target = String(parsedArgs.target ?? parsedArgs.agent_id ?? "");
+					if (target && pendingContinuable.has(target)) pendingContinuable.delete(target);
+				} else if (name === "job_kill") {
+					const jobId = String(parsedArgs.job_id ?? "");
+					if (jobId && pendingJobs.has(jobId)) pendingJobs.delete(jobId);
+				}
+			}
+		}
+	}
+	return pendingContinuable.size > 0 || pendingJobs.size > 0 || anonymousPendingCount > 0;
+}
+function parseArgumentsObject(value) {
+	if (typeof value === "string") try {
+		const parsed = JSON.parse(value);
+		return typeof parsed === "object" && parsed !== null ? parsed : {};
+	} catch {
+		return {};
+	}
+	return typeof value === "object" && value !== null ? value : {};
+}
+const USER_QUESTION_PATTERNS = [
+	/[?？]\s*["'）)』」]*\s*$/u,
+	/(?:请|麻烦您?)(?:告知|确认|指示|选择|提供|决定|回复)/u,
+	/(?:等待|静待|等)(?:您的|你的|您|你|用户)?(?:确认|指示|回复|决定|指令|反馈|选择|输入)/u,
+	/(?:如果您?|如|若)(?:希望|需要|想)(?:继续|执行).*(?:请|告诉我|告知)/u,
+	/(?:是否|要不要|可否)(?:需要我?|继续|同意|允许|采用).*[？?]?/u,
+	/(?:暂时|先|已)?暂停(?:工作|执行|后续).*(?:等|指示|确认|决定|用户)/u,
+	/\b(?:please\s+(?:confirm|let\s+me\s+know|advise|choose|select|provide|indicate|tell\s+me|reply))\b/iu,
+	/\b(?:waiting\s+for|awaiting)\s+(?:your\s+|user\s+)?(?:input|instructions?|reply|response|confirmation|decision|guidance|feedback)\b/iu,
+	/\b(?:would\s+you\s+like|should\s+i|do\s+you\s+want\s+me\s+to|how\s+would\s+you\s+like|which\s+(?:option|approach|strategy)\s+do\s+you\s+prefer)\b/iu,
+	/\b(?:let\s+me\s+know\s+(?:how|what|if|whether|when))\b/iu,
+	/\b(?:paused?\s+(?:work|execution|here|for\s+now)|stopping\s+here)\b/iu
+];
+function isAwaitingUserText(text) {
+	const trimmed = text.trim();
+	if (!trimmed) return false;
+	const lastLine = trimmed.split("\n").filter((l) => l.trim()).at(-1) ?? "";
+	return USER_QUESTION_PATTERNS.some((p) => p.test(lastLine) || p.test(trimmed));
+}
+function narrativeText(blocks) {
+	if (typeof blocks === "string") return blocks.trim();
+	if (!Array.isArray(blocks)) return "";
+	const parts = [];
+	for (const block of blocks) if (block?.type === "text" && typeof block.text === "string") parts.push(block.text);
+	else if (block?.type === "tool-result") parts.push(blockText(block.content));
+	return parts.join("\n").trim();
+}
+/**
+* Whether the agent has paused to ask the user a question, obtain confirmation, or await user instructions.
+*
+* During task execution, an agent may legitimately pause to ask the operator a question
+* (e.g. calling `ask_user_question`, pausing/blocking a goal, or concluding a turn with prose
+* awaiting user guidance). Gating or steering in this state forces the model to keep executing,
+* overriding the user interaction boundary and locking the user out of providing guidance.
+*
+* Returns undefined when the work has already reached its delivery phase (all todos completed
+* with verification evidence), because in that state the agent is delivering the task rather than
+* pausing for input.
+*/
+function inspectUserInteractionPause(events, taskStartSeq, currentTurn) {
+	const delivery = inspectDeliveryPhase(events);
+	if (delivery?.todosComplete && delivery.verification !== void 0) return;
+	let turnEvents = events;
+	if (currentTurn !== void 0) {
+		const turnStartIndex = events.findLastIndex((e) => e.type === "turn/start" && e.data.turn === currentTurn);
+		if (turnStartIndex >= 0) turnEvents = events.slice(turnStartIndex);
+		else {
+			const byTurn = events.filter((e) => e.data?.turn === currentTurn);
+			if (byTurn.length > 0) turnEvents = byTurn;
+		}
+	} else {
+		const lastTurnStart = events.findLastIndex((e) => e.type === "turn/start");
+		if (lastTurnStart >= 0) turnEvents = events.slice(lastTurnStart);
+	}
+	let lastAskUserSeq = -1;
+	for (const event of turnEvents) if (event.type === "tool/call") {
+		if (event.data.name === "ask_user_question") lastAskUserSeq = Math.max(lastAskUserSeq, event.seq);
+	} else if (CODE_DISPATCH_TYPES.has(event.type)) {
+		if (event.data?.name === "ask_user_question") lastAskUserSeq = Math.max(lastAskUserSeq, event.seq);
+	}
+	if (lastAskUserSeq >= 0) {
+		if (!turnEvents.some((event) => {
+			if (event.seq <= lastAskUserSeq) return false;
+			if (event.type === "tool/call") return isConsequential(event.data.name);
+			if (CODE_DISPATCH_TYPES.has(event.type)) {
+				const data = event.data;
+				return data ? isConsequential(data.name) : false;
+			}
+			return false;
+		})) return {
+			paused: true,
+			reason: "ask_user_question in current turn"
+		};
+	}
+	const taskEvents = events.filter((e) => e.seq >= taskStartSeq);
+	let goalPhase;
+	for (const event of taskEvents) if (event.type === "tool/call") {
+		const call = event;
+		if (call.data.name === "update_goal") {
+			const args = parseArgumentsObject(call.data.arguments);
+			if (args.action === "pause") goalPhase = "paused";
+			else if (args.action === "blocked") goalPhase = "blocked";
+			else if (args.action === "resume") goalPhase = "active";
+			else if (args.action === "complete") goalPhase = "complete";
+		}
+	} else if (CODE_DISPATCH_TYPES.has(event.type)) {
+		const data = event.data;
+		if (data?.name === "update_goal") {
+			const args = parseArgumentsObject(data.arguments);
+			if (args.action === "pause") goalPhase = "paused";
+			else if (args.action === "blocked") goalPhase = "blocked";
+			else if (args.action === "resume") goalPhase = "active";
+			else if (args.action === "complete") goalPhase = "complete";
+		}
+	}
+	if (goalPhase === "paused" || goalPhase === "blocked") return {
+		paused: true,
+		reason: "goal is " + goalPhase
+	};
+	const lastAssistant = turnEvents.filter((e) => e.type === "assistant/message").at(-1);
+	if (lastAssistant && lastAssistant.type === "assistant/message") {
+		const content = lastAssistant.data.message?.content ?? [];
+		if (!content.some((b) => b.type === "tool-call")) {
+			const text = narrativeText(content);
+			if (text && isAwaitingUserText(text)) return {
+				paused: true,
+				reason: "assistant awaiting user instructions"
+			};
+		}
+	}
+}
 function analyzeAutoTask(events, policy, sessionId) {
 	const taskStartSeq = latestDirectUserSeq(events);
 	if (taskStartSeq === void 0) return {
@@ -1892,6 +2123,8 @@ function analyzeAutoTask(events, policy, sessionId) {
 		consequentialToolCalls: 0,
 		hasManualSessionVerification: false,
 		manualVerificationAccepted: false,
+		pendingSubagents: false,
+		pendingUserInteraction: false,
 		eligible: false,
 		reason: "no-direct-user-task"
 	};
@@ -1949,6 +2182,8 @@ function analyzeAutoTask(events, policy, sessionId) {
 			criteria: verdict.criteria
 		}, policy.threshold) && !stale(verdict.toSeq);
 	});
+	const pendingSubagents = hasPendingSubagents(events, taskStartSeq);
+	const pendingUserInteraction = inspectUserInteractionPause(events, taskStartSeq) !== void 0;
 	if (policy.mode === "manual") return {
 		taskStartSeq,
 		toolCalls,
@@ -1956,6 +2191,8 @@ function analyzeAutoTask(events, policy, sessionId) {
 		consequentialToolCalls,
 		hasManualSessionVerification,
 		manualVerificationAccepted,
+		pendingSubagents,
+		pendingUserInteraction,
 		eligible: false,
 		reason: "manual-mode"
 	};
@@ -1966,9 +2203,35 @@ function analyzeAutoTask(events, policy, sessionId) {
 		consequentialToolCalls,
 		hasManualSessionVerification,
 		manualVerificationAccepted,
+		pendingSubagents,
+		pendingUserInteraction,
 		eligible: false,
 		reason: "already-verified"
 	};
+	if (pendingSubagents) return {
+		taskStartSeq,
+		toolCalls,
+		completedToolResults,
+		consequentialToolCalls,
+		hasManualSessionVerification,
+		manualVerificationAccepted,
+		pendingSubagents: true,
+		pendingUserInteraction,
+		eligible: false,
+		reason: "pending-subagents"
+	};
+	if (pendingUserInteraction) return {
+		taskStartSeq,
+		toolCalls,
+		completedToolResults,
+		consequentialToolCalls,
+		hasManualSessionVerification,
+		manualVerificationAccepted,
+		pendingSubagents,
+		pendingUserInteraction: true,
+		eligible: false,
+		reason: "user-interaction-paused"
+	};
 	if (consequentialToolCalls === 0) return {
 		taskStartSeq,
 		toolCalls,
@@ -1976,6 +2239,8 @@ function analyzeAutoTask(events, policy, sessionId) {
 		consequentialToolCalls,
 		hasManualSessionVerification,
 		manualVerificationAccepted,
+		pendingSubagents,
+		pendingUserInteraction,
 		eligible: false,
 		reason: "no-consequential-work"
 	};
@@ -1986,6 +2251,8 @@ function analyzeAutoTask(events, policy, sessionId) {
 		consequentialToolCalls,
 		hasManualSessionVerification,
 		manualVerificationAccepted,
+		pendingSubagents,
+		pendingUserInteraction,
 		eligible: false,
 		reason: "no-completed-evidence"
 	};
@@ -1996,6 +2263,8 @@ function analyzeAutoTask(events, policy, sessionId) {
 		consequentialToolCalls,
 		hasManualSessionVerification,
 		manualVerificationAccepted,
+		pendingSubagents,
+		pendingUserInteraction,
 		eligible: false,
 		reason: "insufficient-tool-evidence"
 	};
@@ -2006,6 +2275,8 @@ function analyzeAutoTask(events, policy, sessionId) {
 		consequentialToolCalls,
 		hasManualSessionVerification,
 		manualVerificationAccepted,
+		pendingSubagents,
+		pendingUserInteraction,
 		eligible: true,
 		reason: policy.mode + "-eligible"
 	};
@@ -2173,6 +2444,6 @@ function selectRouteFeedbackDetail(candidates, result, maxChars = MAX_ROUTE_FEED
 	].join("\n"));
 }
 //#endregion
-export { verificationFailed as A, latestDirectUserSeq as C, semanticDecision as D, routedRepeats as E, ScoreCache as F, SingleFlight as I, resolveCacheFile as L, extractSession as M, sanitizeVerifierText as N, semanticReferencesVisible as O, sessionEvents as P, stableHash as R, itemBudget as S, parseSemanticRoute as T, buildSemanticRoutePrompt as _, compareRouteFeedbackDetail as a, inspectDeliveryPhase as b, selectRouteFeedbackDetail as c, AutoVerifierRouter as d, MAX_ROUTED_CHECKPOINTS as f, buildEvidenceIndex as g, boundDecision as h, automaticFeedback as i, verificationVerdict as j, semanticRouteHint as k, sessionAccepted as l, analyzeStructuredRoute as m, PROPOSAL_FEEDBACK_NOTE as n, failedAcceptanceCriteria as o, RECOVERY_FAILURE_CONTEXT_CHARS as p, analyzeAutoTask as r, isSubagentSession as s, MAX_ROUTE_FEEDBACK_CHARS as t, topScoreIndices as u, buildSemanticRouteView as v, nextDiagnosticCycleId as w, inspectRecoverySignal as x, estimateRoutedCalls as y };
+export { semanticDecision as A, resolveCacheFile as B, inspectDeliveryPhase as C, nextDiagnosticCycleId as D, latestDirectUserSeq as E, extractSession as F, sanitizeVerifierText as I, sessionEvents as L, semanticRouteHint as M, verificationFailed as N, parseSemanticRoute as O, verificationVerdict as P, ScoreCache as R, estimateRoutedCalls as S, itemBudget as T, stableHash as V, analyzeStructuredRoute as _, compareRouteFeedbackDetail as a, buildSemanticRoutePrompt as b, inspectUserInteractionPause as c, selectRouteFeedbackDetail as d, sessionAccepted as f, RECOVERY_FAILURE_CONTEXT_CHARS as g, MAX_ROUTED_CHECKPOINTS as h, automaticFeedback as i, semanticReferencesVisible as j, routedRepeats as k, isAwaitingUserText as l, AutoVerifierRouter as m, PROPOSAL_FEEDBACK_NOTE as n, failedAcceptanceCriteria as o, topScoreIndices as p, analyzeAutoTask as r, hasPendingSubagents as s, MAX_ROUTE_FEEDBACK_CHARS as t, isSubagentSession as u, boundDecision as v, inspectRecoverySignal as w, buildSemanticRouteView as x, buildEvidenceIndex as y, SingleFlight as z };
 
-//# sourceMappingURL=auto-DkUMv8FV.js.map
+//# sourceMappingURL=auto-C9YG0E22.js.map