@lazyingart/agintiflow 0.20.207 → 0.20.208

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lazyingart/agintiflow",
-  "version": "0.20.207",
+  "version": "0.20.208",
   "type": "module",
   "description": "AgInTiFlow is a project-aware agent workspace for hybrid wet-dry R&D, hardware-aware intelligence, software automation, and industrial workflows.",
   "license": "Apache-2.0",
@@ -65,6 +65,15 @@
     "scripts/supervision-ledger.js",
     "scripts/smoke-supervision-ledger.js",
     "scripts/smoke-integration-production-mount.js",
+    "scripts/smoke-integration-storage-authority.js",
+    "scripts/smoke-integration-retained-durable-common.js",
+    "scripts/smoke-integration-retained-file-lock.js",
+    "scripts/smoke-integration-retained-event-ledger.js",
+    "scripts/smoke-integration-retained-event-ledger-bundle.js",
+    "scripts/smoke-integration-retained-repository-kernel.js",
+    "scripts/smoke-integration-retained-session-state-store.js",
+    "scripts/smoke-integration-retained-native-session-repository-state.js",
+    "scripts/smoke-integration-runtime-repository-contract.js",
     "scripts/smoke-auxiliary-tools.js",
     "scripts/smoke-agentlink.js",
     "scripts/smoke-auth.js",
@@ -130,7 +139,7 @@
   "scripts": {
     "start": "node run.js",
     "web": "node web.js",
-    "check": "node --check run.js && node --check web.js && node --check bin/aginti-cli.js && node --check bin/aginti-integration.js && node --check bin/aginti-public-research.js && node --check bin/aginti-safe-chat.js && node --check src/*.js && node --check src/mcp/*.js && node --check public/*.js && node --check scripts/postinstall-webapp.js && node --check scripts/seed-supervised-homework.js && node --check scripts/supervision-ledger.js && node --check scripts/smoke-supervision-ledger.js && node --check scripts/smoke-integration-production-mount.js && node --check scripts/eval-provider-attribution.js && node --check scripts/smoke-agentlink.js && node --check scripts/smoke-execution-policy.js && node --check scripts/smoke-math-rendering.js && node --check scripts/smoke-mcp.js && node --check scripts/smoke-public-research-wrapper.js && node --check scripts/smoke-runtime-core.js && node --check scripts/smoke-safe-chat.js && node --check scripts/smoke-web-ui.js && node --check scripts/smoke-scs-evidence-visibility.js && node --check scripts/fixtures/mcp-stdio-smoke-server.mjs",
+    "check": "node --check run.js && node --check web.js && node --check bin/aginti-cli.js && node --check bin/aginti-integration.js && node --check bin/aginti-public-research.js && node --check bin/aginti-safe-chat.js && node --check src/*.js && node --check src/mcp/*.js && node --check public/*.js && node --check scripts/postinstall-webapp.js && node --check scripts/seed-supervised-homework.js && node --check scripts/supervision-ledger.js && node --check scripts/smoke-supervision-ledger.js && node --check scripts/smoke-integration-production-mount.js && node --check scripts/smoke-integration-storage-authority.js && node --check scripts/smoke-integration-retained-durable-common.js && node --check scripts/smoke-integration-retained-file-lock.js && node --check scripts/smoke-integration-retained-event-ledger.js && node --check scripts/smoke-integration-retained-event-ledger-bundle.js && node --check scripts/smoke-integration-retained-repository-kernel.js && node --check scripts/smoke-integration-retained-session-state-store.js && node --check scripts/smoke-integration-retained-native-session-repository-state.js && node --check scripts/smoke-integration-runtime-repository-contract.js && node --check scripts/eval-provider-attribution.js && node --check scripts/smoke-agentlink.js && node --check scripts/smoke-execution-policy.js && node --check scripts/smoke-math-rendering.js && node --check scripts/smoke-mcp.js && node --check scripts/smoke-public-research-wrapper.js && node --check scripts/smoke-runtime-core.js && node --check scripts/smoke-safe-chat.js && node --check scripts/smoke-web-ui.js && node --check scripts/smoke-scs-evidence-visibility.js && node --check scripts/fixtures/mcp-stdio-smoke-server.mjs",
     "setup:toolchain-docker": "scripts/setup-agent-toolchain-docker.sh",
     "smoke:coding-tools": "node scripts/smoke-coding-tools.js",
     "smoke:dynamic-step-budget": "node scripts/smoke-dynamic-step-budget.js",
@@ -186,10 +195,19 @@
     "supervision:ledger": "node scripts/supervision-ledger.js",
     "smoke:supervision-ledger": "node scripts/smoke-supervision-ledger.js",
     "smoke:integration-production-mount": "node scripts/smoke-integration-production-mount.js",
+    "smoke:integration-storage-authority": "node scripts/smoke-integration-storage-authority.js",
+    "smoke:integration-retained-durable-common": "node scripts/smoke-integration-retained-durable-common.js",
+    "smoke:integration-retained-file-lock": "node scripts/smoke-integration-retained-file-lock.js",
+    "smoke:integration-retained-event-ledger": "node scripts/smoke-integration-retained-event-ledger.js",
+    "smoke:integration-retained-event-ledger-bundle": "node --experimental-test-module-mocks scripts/smoke-integration-retained-event-ledger-bundle.js",
+    "smoke:integration-retained-repository-kernel": "node --experimental-test-module-mocks scripts/smoke-integration-retained-repository-kernel.js",
+    "smoke:integration-retained-session-state-store": "node --experimental-test-module-mocks scripts/smoke-integration-retained-session-state-store.js",
+    "smoke:integration-retained-native-session-repository-state": "node --experimental-test-module-mocks scripts/smoke-integration-retained-native-session-repository-state.js",
+    "smoke:integration-runtime-repository-contract": "node scripts/smoke-integration-runtime-repository-contract.js",
     "storage:migrate": "node bin/aginti-cli.js storage migrate",
     "publish:env": "node scripts/npm-publish-from-env.js publish --access public",
     "publish:env:whoami": "node scripts/npm-publish-from-env.js whoami",
-    "test": "npm run check && npm run smoke:integration-production-mount && npm run smoke:localllm-provider && npm run smoke:localllm-model-tiers && npm run smoke:localllm-code-routing && npm run smoke:local-failure-recovery && npm run smoke:localllm-auto-max && npm run smoke:local-resource-policy && npm run smoke:context-budget-recovery && npm run smoke:session-runtime && npm run smoke:runtime-core && npm run smoke:progressive-tools && npm run smoke:truthful-completion && npm run smoke:writing-specialist-routing && npm run eval:local-first-agent && npm run eval:provider-attribution && npm run smoke:runtime-compat && npm run smoke:autoupdate && npm run smoke:web-api && npm run smoke:math-rendering && npm run smoke:web-ui && npm run smoke:web-autostart && npm run smoke:webapp-command && npm run smoke:web-port-fallback && npm run smoke:docker-command && npm run smoke:coding-tools && npm run smoke:dynamic-step-budget && npm run smoke:execution-policy && npm run smoke:aaps-adapter && npm run smoke:auxiliary-tools && npm run smoke:perception-research && npm run smoke:deep-research && npm run smoke:public-research && npm run smoke:safe-chat && npm run smoke:auth && npm run smoke:agentlink && npm run smoke:canvas-artifacts && npm run smoke:capabilities && npm run smoke:mcp && npm run smoke:model-roles && npm run smoke:platform && npm run smoke:permission-modes && npm run smoke:skills && npm run smoke:skillmesh && npm run smoke:supervision-ledger && npm run smoke:tmux-tools && npm run smoke:long-jobs && npm run smoke:run-stdin && npm run smoke:cli-chat && npm run smoke:inbox",
+    "test": "npm run check && npm run smoke:integration-production-mount && npm run smoke:integration-storage-authority && npm run smoke:integration-retained-durable-common && npm run smoke:integration-retained-file-lock && npm run smoke:integration-retained-event-ledger && npm run smoke:integration-retained-event-ledger-bundle && npm run smoke:integration-retained-repository-kernel && npm run smoke:integration-retained-session-state-store && npm run smoke:integration-retained-native-session-repository-state && npm run smoke:integration-runtime-repository-contract && npm run smoke:localllm-provider && npm run smoke:localllm-model-tiers && npm run smoke:localllm-code-routing && npm run smoke:local-failure-recovery && npm run smoke:localllm-auto-max && npm run smoke:local-resource-policy && npm run smoke:context-budget-recovery && npm run smoke:session-runtime && npm run smoke:runtime-core && npm run smoke:progressive-tools && npm run smoke:truthful-completion && npm run smoke:writing-specialist-routing && npm run eval:local-first-agent && npm run eval:provider-attribution && npm run smoke:runtime-compat && npm run smoke:autoupdate && npm run smoke:web-api && npm run smoke:math-rendering && npm run smoke:web-ui && npm run smoke:web-autostart && npm run smoke:webapp-command && npm run smoke:web-port-fallback && npm run smoke:docker-command && npm run smoke:coding-tools && npm run smoke:dynamic-step-budget && npm run smoke:execution-policy && npm run smoke:aaps-adapter && npm run smoke:auxiliary-tools && npm run smoke:perception-research && npm run smoke:deep-research && npm run smoke:public-research && npm run smoke:safe-chat && npm run smoke:auth && npm run smoke:agentlink && npm run smoke:canvas-artifacts && npm run smoke:capabilities && npm run smoke:mcp && npm run smoke:model-roles && npm run smoke:platform && npm run smoke:permission-modes && npm run smoke:skills && npm run smoke:skillmesh && npm run smoke:supervision-ledger && npm run smoke:tmux-tools && npm run smoke:long-jobs && npm run smoke:run-stdin && npm run smoke:cli-chat && npm run smoke:inbox",
     "pack:dry-run": "npm pack --dry-run",
     "smoke:capabilities": "node scripts/smoke-capabilities.js"
   },

package/scripts/smoke-coding-tools.js

@@ -7,6 +7,7 @@ import {
   buildModelTimeoutRetryMessages,
   modelTimeoutRetryRoute,
   repairModelMessageHistory,
+  shouldResetStaticDiscoveryPhase,
   runAgent,
   sanitizeToolResult,
   toolResultForModel,
@@ -18,6 +19,7 @@ import { formatBehaviorContractForPrompt } from "../src/behavior-contract.js";
 import { resolveRuntimeConfig } from "../src/config.js";
 import { readCodebaseMap } from "../src/codebase-map.js";
 import { evaluateCommandPolicy } from "../src/command-policy.js";
+import { shouldReviewToolResult } from "../src/scs-controller.js";
 import {
   engineeringGuidanceForTask,
   recommendedMaxStepsForTask,
@@ -164,6 +166,74 @@ try {
     interruptedDeepSeekState.messages.at(-1)?.content === "Continue with this new request: /review",
     "interrupted repair dropped the new user request"
   );
+  const deepSeekCompactionState = {
+    goal: "Continue a data repair from retained evidence.",
+    plan: "Use the verified source evidence, repair once, then test.",
+    stepsCompleted: 4,
+    meta: {},
+    messages: [
+      { role: "system", content: "system" },
+      { role: "user", content: "inspect the exact source" },
+      {
+        role: "assistant",
+        content: "",
+        reasoning_content: "The source must be read before editing.",
+        tool_calls: [
+          { id: "deep-read", type: "function", function: { name: "read_file", arguments: '{"path":"analysis.py"}' } },
+        ],
+      },
+      { role: "tool", tool_call_id: "deep-read", content: '{"ok":true,"path":"analysis.py","content":"verified source"}' },
+    ],
+  };
+  const deepSeekCompacted = buildModelTimeoutRetryMessages(
+    deepSeekCompactionState,
+    { provider: "deepseek", model: "deepseek-v4-pro", contextWindowTokens: 32768 },
+    { url: "", title: "No browser page open" },
+    5,
+    new Error("synthetic timeout")
+  );
+  assert(
+    !deepSeekCompacted.some((message) => message.role === "assistant" && Array.isArray(message.tool_calls)),
+    "DeepSeek compaction synthesized assistant tool calls without original reasoning_content"
+  );
+  assert(
+    deepSeekCompacted.some(
+      (message) => message.role === "user" && /Retained runtime tool evidence/.test(message.content) && /analysis\.py/.test(message.content) && /verified source/.test(message.content)
+    ),
+    "DeepSeek compaction dropped bounded source evidence while removing synthetic tool-call messages"
+  );
+  assert(
+    !shouldReviewToolResult(
+      { ok: true, toolName: "read_file", path: "analysis.py" },
+      { meta: { toolLoop: { warned: ["old-read"], recent: [{ toolName: "read_file", ok: false }] } } }
+    ),
+    "SCS scheduled a redundant review for a successful read because an older read failed"
+  );
+  assert(
+    shouldReviewToolResult(
+      { ok: false, blocked: true, toolName: "read_file", reason: "exact read blocked" },
+      { meta: { toolLoop: { warned: [], recent: [] } } }
+    ),
+    "SCS stopped reviewing an exact blocked tool result"
+  );
+  assert(
+    !shouldResetStaticDiscoveryPhase({
+      ok: true,
+      toolName: "run_command",
+      args: { command: 'echo "SOURCE"; cat analysis.py; echo "DIFF"; git diff -- analysis.py' },
+      commandPolicy: { writesWorkspace: false },
+    }),
+    "composite read-only shell discovery incorrectly reset the bounded discovery phase"
+  );
+  assert(
+    shouldResetStaticDiscoveryPhase({
+      ok: true,
+      toolName: "run_command",
+      args: { command: "python analysis.py" },
+      commandPolicy: { writesWorkspace: true },
+    }),
+    "a successful workspace-writing command did not reset the discovery phase"
+  );
   const workspaceToolConfig = {
     commandCwd: workspace,
     allowFileTools: true,
@@ -1111,6 +1181,14 @@ try {
     localTimeoutRoute.model === "localllm-fast" && localTimeoutRoute.retryTimeoutMs === 90000,
     "LocalLLM timeout retry did not switch to its same-boundary fast route"
   );
+  const defaultLocalTimeoutRoute = modelTimeoutRetryRoute({
+    provider: "localllm",
+    model: "localllm-fast",
+  });
+  assert(
+    defaultLocalTimeoutRoute.timeoutMs === 300000 && defaultLocalTimeoutRoute.retryTimeoutMs === 600000,
+    "LocalLLM default timeout did not allow bounded slow local generation and one longer retry"
+  );
   const artifactTimeoutMessages = buildModelTimeoutRetryMessages(
     {
       meta: {

package/scripts/smoke-context-budget-recovery.js

@@ -163,6 +163,10 @@ const runtimeMessages = buildContextBudgetCompactionMessages(
   { reason: "test context recovery" }
 );
 const runtimeText = runtimeMessages.map((message) => message.content || "").join("\n");
+assert.ok(
+  estimateMessageTokens(runtimeMessages) <= 12288,
+  "runtime compaction exceeded the bounded LocalLLM retry target"
+);
 assert.ok(runtimeText.includes(HEAD));
 assert.ok(runtimeText.includes(TAIL));
 assert.ok(runtimeText.includes("latest interruption"));

package/scripts/smoke-dynamic-step-budget.js

@@ -22,6 +22,8 @@ import {
   recordExactOutputProgress,
   recordStaticDiscoveryProgress,
   rememberCompletedDeepResearch,
+  repeatedNoProgressToolBlock,
+  repeatedSuccessfulMutationBlock,
   repeatedStaticToolBlock,
   shouldResetStaticDiscoveryPhase,
 } from "../src/agent-runner.js";
@@ -59,6 +61,39 @@ try {
   assert(normalizeDynamicStepsMode("off") === "off", "dynamic mode off did not normalize");
   assert(normalizeDynamicStepsMode("always") === "on", "dynamic mode always did not normalize to on");
   assert(normalizeDynamicStepsMode("smart") === "auto", "dynamic mode smart did not normalize to auto");
+  const inheritedBudget = createStepBudgetState(
+    { maxSteps: 16, dynamicSteps: "off" },
+    {
+      stepsCompleted: 0,
+      meta: {
+        stepBudget: {
+          initialMaxSteps: 30,
+          currentMaxSteps: 40,
+          extensionsUsed: 1,
+          lastExtensionStep: 28,
+        },
+      },
+    }
+  );
+  assert(inheritedBudget.currentMaxSteps === 40, "ordinary resume did not retain its prior expanded budget");
+  const explicitBudget = createStepBudgetState(
+    { maxSteps: 16, dynamicSteps: "off", resetStepBudget: true },
+    {
+      stepsCompleted: 0,
+      meta: {
+        stepBudget: {
+          initialMaxSteps: 30,
+          currentMaxSteps: 40,
+          extensionsUsed: 1,
+          lastExtensionStep: 28,
+        },
+      },
+    }
+  );
+  assert(explicitBudget.initialMaxSteps === 16, "explicit resumed max-steps did not reset the initial budget");
+  assert(explicitBudget.currentMaxSteps === 16, "a prior extension overrode explicit resumed max-steps");
+  assert(explicitBudget.extensionsUsed === 0, "explicit resumed max-steps retained stale extension usage");
+  assert(explicitBudget.resetFromExplicitOverride, "explicit resumed max-steps reset was not recorded");
   assert(isStaticDiscoveryToolCall("run_command", { command: "ls -la ../Musia" }), "static ls discovery was not classified");
   assert(isStaticDiscoveryToolCall("read_image", { path: "snapshot.png" }), "image perception was not classified as static discovery");
   assert(
@@ -97,6 +132,126 @@ try {
     ) === null,
     "a bounded continuation read was mistaken for an exact reread"
   );
+  const repeatedProbeArgs = { command: "python -c \"print(100.0)\"" };
+  const repeatedProbeSignature = staticToolCallSignature("run_command", repeatedProbeArgs, {
+    commandCwd: workspace,
+  });
+  const repeatedProbeState = {
+    meta: {
+      toolLoop: {
+        stagnationEpoch: 4,
+        recent: [
+          {
+            signature: repeatedProbeSignature,
+            toolName: "run_command",
+            ok: true,
+            blocked: false,
+            noProgressProbe: true,
+            outcomeFingerprint: "same-output",
+            stagnationEpoch: 4,
+          },
+          {
+            signature: repeatedProbeSignature,
+            toolName: "run_command",
+            ok: true,
+            blocked: false,
+            noProgressProbe: true,
+            outcomeFingerprint: "same-output",
+            stagnationEpoch: 4,
+          },
+        ],
+      },
+    },
+  };
+  assert(
+    repeatedNoProgressToolBlock(repeatedProbeState, "run_command", repeatedProbeArgs, {
+      commandCwd: workspace,
+    })?.category === "repeated-no-progress-call",
+    "a third identical unchanged shell probe was not blocked"
+  );
+  assert(
+    repeatedNoProgressToolBlock(
+      {
+        meta: {
+          toolLoop: {
+            stagnationEpoch: 5,
+            recent: repeatedProbeState.meta.toolLoop.recent,
+          },
+        },
+      },
+      "run_command",
+      repeatedProbeArgs,
+      { commandCwd: workspace }
+    ) === null,
+    "a new mutation epoch did not reopen bounded validation"
+  );
+  assert(
+    repeatedNoProgressToolBlock(
+      repeatedProbeState,
+      "run_command",
+      { command: "python monitor.py --status" },
+      { commandCwd: workspace }
+    ) === null,
+    "an explicit status polling command was incorrectly blocked"
+  );
+  const repeatedPatchArgs = {
+    path: "analysis.py",
+    search: "signal = raw_signal",
+    replace: "signal = raw_signal - offset",
+    searchHash: "search-hash",
+    replaceHash: "replace-hash",
+  };
+  const repeatedPatchSignature = staticToolCallSignature("apply_patch", repeatedPatchArgs, {
+    commandCwd: workspace,
+  });
+  const repeatedPatchState = {
+    meta: {
+      toolLoop: {
+        stagnationEpoch: 7,
+        recent: [
+          {
+            signature: repeatedPatchSignature,
+            toolName: "apply_patch",
+            ok: true,
+            blocked: false,
+            successfulMutation: true,
+            stagnationEpoch: 7,
+          },
+        ],
+      },
+    },
+  };
+  assert(
+    repeatedSuccessfulMutationBlock(repeatedPatchState, "apply_patch", repeatedPatchArgs, {
+      commandCwd: workspace,
+    })?.category === "repeated-successful-mutation",
+    "an exact already-successful patch was not blocked in the same mutation epoch"
+  );
+  assert(
+    repeatedSuccessfulMutationBlock(
+      {
+        meta: {
+          toolLoop: {
+            stagnationEpoch: 8,
+            recent: repeatedPatchState.meta.toolLoop.recent,
+          },
+        },
+      },
+      "apply_patch",
+      repeatedPatchArgs,
+      { commandCwd: workspace }
+    ) === null,
+    "a user continuation did not reopen an intentional exact patch"
+  );
+  assert(
+    repeatedSuccessfulMutationBlock(
+      repeatedPatchState,
+      "apply_patch",
+      { ...repeatedPatchArgs, replaceHash: "different-replacement-hash" },
+      { commandCwd: workspace }
+    ) === null,
+    "a materially different patch was mistaken for an exact replay"
+  );
   const completedResearchState = {
     goal: "Create one cited report",
     meta: { goalContract: { revision: 1, currentHash: "goal-one" } },
@@ -640,6 +795,24 @@ try {
   const state = await store.loadState();
   assert(state.meta?.stepBudget?.extensionsUsed === 1, "dynamic budget state did not persist extension count");
 
+  const resumed = await runAgent({
+    ...config,
+    goal: "Create notes/explicit-resume-budget.md with one concise line.",
+    resume: run.sessionId,
+    sessionId: run.sessionId,
+    maxSteps: 2,
+    dynamicSteps: "off",
+    runtimePatch: { maxSteps: 2, dynamicSteps: "off" },
+    expectedRuntimeRevision: state.meta?.runtimeConfig?.revision,
+  });
+  assert(!resumed.stopped, "explicit bounded resume stopped before completing its simple task");
+  const resumedEvents = await store.loadEvents();
+  const resumedBudget = resumedEvents.filter((event) => event.type === "budget.initialized").at(-1)?.data;
+  assert(resumedBudget?.initialMaxSteps === 2, "runAgent did not apply explicit resumed max-steps as the initial budget");
+  assert(resumedBudget?.currentMaxSteps === 2, "runAgent inherited an older expanded budget over explicit resumed max-steps");
+  assert(resumedBudget?.extensionsUsed === 0, "runAgent retained stale extension usage after an explicit resumed max-steps patch");
+  assert(resumedBudget?.resetFromExplicitOverride === true, "runAgent did not record the explicit budget reset boundary");
+
   await fs.rm(tempRoot, { recursive: true, force: true });
   console.log("smoke-dynamic-step-budget ok");
 } catch (error) {