feat(agents): rename scheduler agent tool cron -> automations (#114841)

* refactor(agents): route scheduler tool-name consumers through canonical identity Introduce AUTOMATIONS_TOOL_NAME + isAutomationsToolName() in src/agents/tools/automations-tool-name.ts as the single source of truth for the scheduler agent tool's name, and convert every exact-name consumer: factory descriptors, deferred-followup availability detection, add-counting, policy deny lists, mutation classification, trusted media set, tool catalog id, system-prompt tool order and tool-line map, sandbox deny defaults, delegation capability map, MCP loopback probes, and local-model lean deny. Behavior-neutral: the constant still resolves to "cron". Prepares the rename in RFC openclaw/rfcs#50 so the flip is a one-line change with no scattered literals. * feat(agents): rename scheduler agent tool cron -> automations Flip AUTOMATIONS_TOOL_NAME to "automations" and register the legacy name: - TOOL_NAME_ALIASES gains cron -> automations, so persisted toolsAllow/ toolsDeny lists, tool groups, and creator allowlists written before the rename keep matching through the same shipped mechanism as bash -> exec. No doctor rewrite needed. - isAutomationsToolName() accepts legacy names so saved transcripts keep their mutation/replay-safety classification; MUTATING_TOOL_NAMES retains the legacy entry for the same reason. - Tool label, catalog label, and tool-search keywords follow the rename ("cron" kept as a search synonym). - Regression tests cover old-name policy matching (allow and deny), legacy transcript replay classification, and legacy creator allowlists normalizing to the canonical id. Model-facing description strings still say cron; those move in the follow-up strings PR. Part of RFC openclaw/rfcs#50 Phase 1. * test(agents): update creator-cap expectations for canonical automations id The creator tool surface derives from normalized live tool names, so derived toolsAllow outputs now emit "automations". Passthrough paths without a creator cap keep storing user input verbatim; those expectations stay on the legacy name as stored-data coverage. * fix(gateway): canonicalize legacy cron tool calls and restore scheduler deny protection Review follow-ups from ClawSweeper and Codex on the rename (RFC 0026): - MCP loopback tools/call resolves legacy "cron" names to the published automations tool without re-advertising the old name in tools/list. - Gateway /tools/invoke canonicalizes legacy names before core-id checks and exact-name dispatch, so pre-rename integrations keep working. - Security fix: dangerous-tools deny lists (owner-only HTTP deny and control-plane set) were keyed on the literal "cron", so the renamed tool silently lost default-deny and owner-only protection on the HTTP invoke surface. Lists now use the canonical constant, and the gateway.tools.allow un-deny filter normalizes both sides so legacy allow entries still lift it. - Voice high-impact confirmation list and MCP serve creator allowlist follow the canonical name. Existing cron-regression suite now proves the legacy path end to end: default deny 404 for both names, legacy allow entry lifts the deny, and non-owner protection holds. * fix(agents): cover stdio MCP legacy calls, probe prompts, and prompt snapshots for the rename - stdio MCP servers (openclaw-tools-serve / plugin tools handlers) resolve legacy "cron" callTool names to the published canonical tool, matching the HTTP loopback behavior; listTools stays canonical-only. - Live probe prompts instruct harnesses to load/call the automations MCP tool (mcp__openclaw__automations) instead of the retired name. - Prompt snapshot fixture filter follows the canonical name (the renamed tool had silently dropped out of the Codex dynamic-tools snapshots); snapshots regenerated as a clean rename. - Type-cast the new mcp-http handler test payloads for check-test-types. * test(agents): update tool-surface expectations for the automations rename CI-surfaced fallout in shards not covered by the focused local runs: tool availability, agent-config filtering, coding-tools construction, model-provider lean policy, and skill dispatch all assert the scheduler tool's surface name. Mock fixtures and expectations follow the canonical id; legacy-name coverage stays in the dedicated policy/creator-cap/invoke regression suites. * test(gateway): update tool-resolution exclude expectations for automations rename * test(security): update trust-model audit expectations for automations rename * docs(agents): declare cron a permanent scheduler-tool alias per owner decision Maintainer decision (Omar): cron is not being retired anywhere — config keys, RPC methods, schedule syntax, and the CLI token all keep it, and the tool alias follows the same permanent contract as bash -> exec. No doctor rewrite and no removal window; comments updated to state the contract instead of a deprecation plan. * chore(agents): regen prompt snapshots after rebase onto main * fix(agents): teach canonical automations tool in fallback guidance and reuse the identity source Review follow-ups: the structured-list fallback still taught models the cron tool; the cron-scope test echoed its own stub; MCP serve allowlist and voice confirmation hardcoded the name instead of the canonical constant. * fix(mcp): place automations identity import outside the header comment --------- Co-authored-by: Omar Shahine <10343873+omarshahine@users.noreply.github.com>
2026-08-12 21:53:00 -06:00 · 2026-07-30 07:16:13 -07:00
parent 319fd692d1
commit 364d8be010
56 changed files with 323 additions and 155 deletions
@@ -200,7 +200,7 @@ describe("Agent-specific tool filtering", () => {
    const toolNames = tools.map((t) => t.name);
    expect(toolNames).toContain("read");
    expect(toolNames).not.toContain("browser");
-    expect(toolNames).not.toContain("cron");
+    expect(toolNames).not.toContain("automations");
    expect(toolNames).not.toContain("message");
  });

@@ -502,7 +502,7 @@ describe("Agent-specific tool filtering", () => {

    expect(ownerTools).toContain("exec");
    expect(ownerTools).toContain("process");
-    expect(ownerTools).toContain("cron");
+    expect(ownerTools).toContain("automations");
    expect(ownerTools).toContain("gateway");
    expect(ownerTools).toContain("nodes");
    expect(ownerTools).toContain("openclaw");
@@ -511,7 +511,7 @@ describe("Agent-specific tool filtering", () => {
    expect(ownerTools).toContain("conversations_turn");
    expect(nonOwnerTools).not.toContain("exec");
    expect(nonOwnerTools).not.toContain("process");
-    expect(nonOwnerTools).not.toContain("cron");
+    expect(nonOwnerTools).not.toContain("automations");
    expect(nonOwnerTools).not.toContain("gateway");
    expect(nonOwnerTools).not.toContain("nodes");
    expect(nonOwnerTools).not.toContain("openclaw");
@@ -28,7 +28,7 @@ describe("tool availability", () => {
    const tools = createOpenClawCodingTools();
    const toolNames = tools.map((tool) => tool.name);
    expect(toolNames).toContain("plugin_login");
-    expect(toolNames).toContain("cron");
+    expect(toolNames).toContain("automations");
    expect(toolNames).toContain("gateway");
    expect(toolNames).toContain("nodes");
    expect(toolNames).toContain("openclaw");
@@ -422,7 +422,7 @@ describe("createOpenClawCodingTools", () => {
    });
    const names = new Set(tools.map((tool) => tool.name));

-    expect(names.has("cron")).toBe(true);
+    expect(names.has("automations")).toBe(true);
    expect(names.has("gateway")).toBe(true);
    expect(names.has("nodes")).toBe(true);
  });
@@ -432,13 +432,13 @@ describe("createOpenClawCodingTools", () => {
      createOpenClawCodingTools({
        config: testConfig,
      }),
-      ["cron"],
+      ["automations"],
    );

-    expect(allowed.map((tool) => tool.name)).toEqual(["cron"]);
+    expect(allowed.map((tool) => tool.name)).toEqual(["automations"]);
    expect(
      buildEmptyExplicitToolAllowlistError({
-        sources: [{ label: "runtime toolsAllow", entries: ["cron"] }],
+        sources: [{ label: "runtime toolsAllow", entries: ["automations"] }],
        callableToolNames: allowed.map((tool) => tool.name),
        toolsEnabled: true,
      }),
@@ -1253,12 +1253,12 @@ describe("createOpenClawCodingTools", () => {
    createOpenClawCodingTools({
      sessionKey: "agent:main:whatsapp:group:restricted-room",
      config: {
-        tools: { allow: ["read", "exec", "process", "cron"] },
+        tools: { allow: ["read", "exec", "process", "automations"] },
        channels: {
          whatsapp: {
            groups: {
              "restricted-room": {
-                tools: { allow: ["read", "cron"] },
+                tools: { allow: ["read", "automations"] },
              },
            },
          },
@@ -1269,7 +1269,7 @@ describe("createOpenClawCodingTools", () => {
    expect(createOpenClawToolsMock).toHaveBeenCalledTimes(1);
    const cronAllow = latestCreateOpenClawToolsOptions().cronCreatorToolAllowlist;
    const cronAllowNames = cronCreatorToolNames(cronAllow);
-    expectListIncludes(cronAllowNames, ["read", "cron"]);
+    expectListIncludes(cronAllowNames, ["read", "automations"]);
    expect(cronAllowNames?.includes("exec")).toBe(false);
    expect(cronAllowNames?.includes("process")).toBe(false);
  });
@@ -1284,7 +1284,7 @@ describe("createOpenClawCodingTools", () => {
    const cronAllowNames = cronCreatorToolNames(
      latestCreateOpenClawToolsOptions().cronCreatorToolAllowlist,
    );
-    expectListIncludes(cronAllowNames, ["read", "cron", "exec"]);
+    expectListIncludes(cronAllowNames, ["read", "automations", "exec"]);
  });

  it("lets embedded attempts refresh a caller-owned cron creator tool surface", () => {
@@ -1295,22 +1295,22 @@ describe("createOpenClawCodingTools", () => {
    > = [];

    createOpenClawCodingTools({
-      config: { tools: { allow: ["read", "cron"] } },
+      config: { tools: { allow: ["read", "automations"] } },
      cronCreatorToolAllowlistRef,
    });

    expect(createOpenClawToolsMock).toHaveBeenCalledTimes(1);
    const cronAllow = latestCreateOpenClawToolsOptions().cronCreatorToolAllowlist;
    expect(cronAllow).toBe(cronCreatorToolAllowlistRef);
-    expect(cronCreatorToolNames(cronAllow)).toEqual(["read", "cron"]);
+    expect(cronCreatorToolNames(cronAllow)).toEqual(["read", "automations"]);

    replaceWithEffectiveCronCreatorToolAllowlist(cronCreatorToolAllowlistRef, [
      stubTool("read"),
-      stubTool("cron"),
+      stubTool("automations"),
      stubTool("bundle_mcp_search"),
    ]);

-    expect(cronCreatorToolNames(cronAllow)).toEqual(["read", "cron", "bundle_mcp_search"]);
+    expect(cronCreatorToolNames(cronAllow)).toEqual(["read", "automations", "bundle_mcp_search"]);
  });

  it("passes deny-restricted tool surface to cron-created agent turns", () => {
@@ -1320,7 +1320,7 @@ describe("createOpenClawCodingTools", () => {
    createOpenClawCodingTools({
      sessionKey: "agent:main:whatsapp:group:restricted-room",
      config: {
-        tools: { allow: ["read", "exec", "process", "cron"] },
+        tools: { allow: ["read", "exec", "process", "automations"] },
        channels: {
          whatsapp: {
            groups: {
@@ -1336,7 +1336,7 @@ describe("createOpenClawCodingTools", () => {
    expect(createOpenClawToolsMock).toHaveBeenCalledTimes(1);
    const cronAllow = latestCreateOpenClawToolsOptions().cronCreatorToolAllowlist;
    const cronAllowNames = cronCreatorToolNames(cronAllow);
-    expectListIncludes(cronAllowNames, ["read", "cron"]);
+    expectListIncludes(cronAllowNames, ["read", "automations"]);
    expect(cronAllowNames?.includes("exec")).toBe(false);
    expect(cronAllowNames?.includes("process")).toBe(false);
  });
@@ -1374,7 +1374,7 @@ describe("createOpenClawCodingTools", () => {

  it("preserves action enums in normalized schemas", () => {
    const defaultTools = createOpenClawCodingTools({ config: testConfig });
-    const toolNames = ["canvas", "nodes", "cron", "gateway", "message"];
+    const toolNames = ["canvas", "nodes", "automations", "gateway", "message"];
    const missingNames = toolNames.filter(
      (name) => !defaultTools.some((candidate) => candidate.name === name),
    );
@@ -1691,7 +1691,7 @@ describe("createOpenClawCodingTools", () => {
    expect(names.has("browser")).toBe(true);
    expect(names.has("canvas")).toBe(true);
    expect(names.has("gateway")).toBe(true);
-    expect(names.has("cron")).toBe(true);
+    expect(names.has("automations")).toBe(true);
    expect(names.has("nodes")).toBe(true);
  });

@@ -28,7 +28,7 @@ vi.mock("./openclaw-tools.js", async (importOriginal) => {
  return {
    createOpenClawTools: (options: unknown) => {
      mocks.createOpenClawToolsOptions(options);
-      return [mocks.stubTool("cron")];
+      return [mocks.stubTool(AUTOMATIONS_TOOL_NAME)];
    },
    filterToolsByClientCaps: actual.filterToolsByClientCaps,
  };
@@ -37,6 +37,7 @@ vi.mock("./openclaw-tools.js", async (importOriginal) => {
 import "./test-helpers/fast-bash-tools.js";
 import "./test-helpers/fast-coding-tools.js";
 import { createOpenClawCodingTools } from "./agent-tools.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";

 function firstOpenClawToolsOptions(): { cronSelfRemoveOnlyJobId?: string } | undefined {
  return mocks.createOpenClawToolsOptions.mock.calls[0]?.[0] as
@@ -55,7 +56,7 @@ describe("createOpenClawCodingTools cron scope", () => {
      jobId: "job-current",
    });

-    expect(tools.map((tool) => tool.name)).toContain("cron");
+    expect(tools.map((tool) => tool.name)).toContain(AUTOMATIONS_TOOL_NAME);
    expect(firstOpenClawToolsOptions()?.cronSelfRemoveOnlyJobId).toBe("job-current");
  });

@@ -9,6 +9,7 @@ import { describeExecTool, describeProcessTool } from "./bash-tools.descriptions
 import { copyBeforeToolCallHookMarker } from "./before-tool-call-metadata.js";
 import { copyChannelAgentToolMeta } from "./channel-tools.js";
 import { copyToolTerminalPresentation } from "./tool-terminal-presentation.js";
+import { isAutomationsToolName } from "./tools/automations-tool-name.js";

 function replaceDescription(tool: AnyAgentTool, description: string): AnyAgentTool {
  const updated = { ...tool, description };
@@ -24,7 +25,7 @@ export function applyDeferredFollowupToolDescriptions(
  tools: AnyAgentTool[],
  params?: { agentId?: string },
 ): AnyAgentTool[] {
-  const hasCronTool = tools.some((tool) => tool.name === "cron");
+  const hasCronTool = tools.some((tool) => isAutomationsToolName(tool.name));
  return tools.map((tool) => {
    if (tool.name === "exec") {
      return replaceDescription(tool, describeExecTool({ agentId: params?.agentId, hasCronTool }));
@@ -173,7 +173,7 @@ describe("applyModelProviderToolPolicy", () => {
      [
        { name: "read" },
        { name: "browser" },
-        { name: "cron" },
+        { name: "automations" },
        { name: "message" },
        { name: "exec" },
      ] as unknown as AnyAgentTool[],
@@ -201,7 +201,7 @@ describe("applyModelProviderToolPolicy", () => {
      [
        { name: "read" },
        { name: "browser" },
-        { name: "cron" },
+        { name: "automations" },
        { name: "message" },
        { name: "exec" },
      ] as unknown as AnyAgentTool[],
@@ -233,7 +233,7 @@ describe("applyModelProviderToolPolicy", () => {
      [
        { name: "read" },
        { name: "browser" },
-        { name: "cron" },
+        { name: "automations" },
        { name: "message" },
        { name: "exec" },
      ] as unknown as AnyAgentTool[],
@@ -265,7 +265,7 @@ describe("applyModelProviderToolPolicy", () => {
      [
        { name: "read" },
        { name: "browser" },
-        { name: "cron" },
+        { name: "automations" },
        { name: "message" },
        { name: "exec" },
      ] as unknown as AnyAgentTool[],
@@ -303,7 +303,7 @@ describe("applyModelProviderToolPolicy", () => {
      [
        { name: "read" },
        { name: "browser" },
-        { name: "cron" },
+        { name: "automations" },
        { name: "message" },
        { name: "exec" },
      ] as unknown as AnyAgentTool[],
@@ -332,7 +332,7 @@ describe("applyModelProviderToolPolicy", () => {
      },
    );

-    expect(toolNames(filtered)).toEqual(["read", "browser", "cron", "message", "exec"]);
+    expect(toolNames(filtered)).toEqual(["read", "browser", "automations", "message", "exec"]);
  });

  it("keeps heavyweight tools when the experimental lean local-model flag is not enabled", () => {
@@ -340,7 +340,7 @@ describe("applyModelProviderToolPolicy", () => {
      [
        { name: "read" },
        { name: "browser" },
-        { name: "cron" },
+        { name: "automations" },
        { name: "message" },
        { name: "exec" },
      ] as unknown as AnyAgentTool[],
@@ -360,6 +360,6 @@ describe("applyModelProviderToolPolicy", () => {
      },
    );

-    expect(toolNames(filtered)).toEqual(["read", "browser", "cron", "message", "exec"]);
+    expect(toolNames(filtered)).toEqual(["read", "browser", "automations", "message", "exec"]);
  });
 });
@@ -41,6 +41,7 @@ import {
  normalizeToolName,
  resolveToolProfilePolicy,
 } from "./tool-policy.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";

 export { resolveProviderToolPolicy };

@@ -54,7 +55,7 @@ const SUBAGENT_TOOL_DENY_ALWAYS = [
  "agents_list",
  // Status/scheduling - main agent coordinates
  "session_status",
-  "cron",
+  AUTOMATIONS_TOOL_NAME,
  // Direct session sends - subagents communicate through announce chain
  "sessions_send",
  "conversations_list",
@@ -2,6 +2,8 @@
 * Static identity for names that select core agent factory families before assembly.
 */

+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";
+
 export type CoreToolFactoryFamily = "base-coding" | "shell" | "openclaw";

 type CoreToolFactoryDescriptor = {
@@ -25,7 +27,7 @@ const CORE_TOOL_FACTORY_DESCRIPTORS = [
  { name: "conversations_list", family: "openclaw" },
  { name: "conversations_send", family: "openclaw" },
  { name: "conversations_turn", family: "openclaw" },
-  { name: "cron", family: "openclaw" },
+  { name: AUTOMATIONS_TOOL_NAME, family: "openclaw" },
  { name: "dashboard", family: "openclaw" },
  { name: "gateway", family: "openclaw" },
  { name: "get_goal", family: "openclaw" },
@@ -1,5 +1,6 @@
 import { isCompletionReportInputProvenance } from "../sessions/input-provenance.js";
 import { normalizeToolName } from "./tool-policy.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";
 import type { AnyAgentTool } from "./tools/common.js";
 import { ToolAuthorizationError } from "./tools/common.js";

@@ -14,7 +15,7 @@ const NEW_DELEGATION_TOOL_NAMES = new Set([
 ]);

 const REPORT_ONLY_TOOL_ACTIONS: ReadonlyMap<string, ReadonlySet<string>> = new Map([
-  ["cron", new Set(["get", "list", "remove", "runs", "status"])],
+  [AUTOMATIONS_TOOL_NAME, new Set(["get", "list", "remove", "runs", "status"])],
  ["image_generate", new Set(["list", "status"])],
  ["music_generate", new Set(["list", "status"])],
  ["video_generate", new Set(["list", "status"])],
@@ -106,6 +106,7 @@ import {
  settleAskUserPromptDelivery,
  waitForAskUserPromptReady,
 } from "./tools/ask-user-tool.js";
+import { isAutomationsToolName } from "./tools/automations-tool-name.js";

 type ExecApprovalReplyModule = typeof import("../infra/exec-approval-reply.js");
 type HookRunnerGlobalModule = typeof import("../plugins/hook-runner-global.js");
@@ -1580,7 +1581,7 @@ export async function handleToolExecutionEnd(
  // Track committed reminders only when cron.add completed successfully.
  if (
    !isToolError &&
-    ((toolName === "cron" && isCronAddAction(startArgs)) ||
+    ((isAutomationsToolName(toolName) && isCronAddAction(startArgs)) ||
      (isExecToolName(toolName) && didShellCronAddSucceed(startArgs, result)))
  ) {
    ctx.state.successfulCronAdds += 1;
@@ -34,6 +34,7 @@ import {
  readToolResultDetails,
  readToolResultStatus,
 } from "./tool-result-error.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";

 export { isToolResultError };

@@ -602,7 +603,7 @@ const TRUSTED_TOOL_RESULT_MEDIA = new Set([
  "apply_patch",
  "browser",
  "canvas",
-  "cron",
+  AUTOMATIONS_TOOL_NAME,
  "edit",
  "exec",
  "gateway",
@@ -10,10 +10,11 @@ import { resolveAgentConfig, resolveDefaultAgentId } from "./agent-scope-config.
 import type { AnyAgentTool } from "./agent-tools.types.js";
 import { compileGlobPatterns, matchesAnyGlobPattern } from "./glob-pattern.js";
 import { expandToolGroups, normalizeToolName } from "./tool-policy.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";

 const LOCAL_MODEL_LEAN_DENY_TOOL_NAMES = new Set([
  "browser",
-  "cron",
+  AUTOMATIONS_TOOL_NAME,
  "image_generate",
  "message",
  "music_generate",
@@ -0,0 +1,20 @@
+// @vitest-environment node
+import { describe, expect, it } from "vitest";
+import { buildOpenClawToolFallbackText } from "./prompt-surface.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";
+
+describe("buildOpenClawToolFallbackText", () => {
+  it("teaches the canonical scheduler tool, never the legacy alias", () => {
+    const text = buildOpenClawToolFallbackText({
+      surface: "openclaw_main",
+      execToolName: "exec",
+      processToolName: "process",
+    });
+
+    expect(text).toContain(`- ${AUTOMATIONS_TOOL_NAME}:`);
+    // The fallback is model-facing guidance; it must not teach "cron" as a
+    // tool name even though the inbound alias stays accepted (RFC 0026).
+    expect(text).not.toMatch(/-\s*cron\b/);
+    expect(text.toLowerCase()).not.toContain("cron job");
+  });
+});
@@ -6,6 +6,7 @@
 import { isOpenClawMainPromptSurface } from "../plugins/agent-prompt-surface-kind.js";
 import type { AgentPromptSurfaceKind } from "../plugins/types.js";
 import { isAcpSessionKey, isSubagentSessionKey } from "../routing/session-key.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";

 /** Builds fallback tool guidance when a runtime cannot render the structured tool list. */
 export function buildOpenClawToolFallbackText(params: {
@@ -25,7 +26,7 @@ export function buildOpenClawToolFallbackText(params: {
      "- browser: control OpenClaw's dedicated browser",
      "- canvas: present/eval/snapshot the Canvas",
      "- nodes: list/describe/notify/camera/screen on paired nodes",
-      "- cron: manage cron jobs and wake events (use for reminders; when scheduling a reminder, write the systemEvent text as something that will read like a reminder when it fires, and mention that it is a reminder depending on the time gap between setting and firing; include recent context in reminder text if appropriate)",
+      `- ${AUTOMATIONS_TOOL_NAME}: manage automations (scheduled jobs) and wake events (use for reminders; when scheduling a reminder, write the systemEvent text as something that will read like a reminder when it fires, and mention that it is a reminder depending on the time gap between setting and firing; include recent context in reminder text if appropriate)`,
      "- conversations_list: list exact external conversation addresses",
      "- conversations_send: send directly to an external conversation",
      "- conversations_turn: send and wait for a correlated external reply",
@@ -6,6 +6,7 @@
 import path from "node:path";
 import { CHANNEL_IDS } from "../../channels/ids.js";
 import { STATE_DIR } from "../../config/paths.js";
+import { AUTOMATIONS_TOOL_NAME } from "../tools/automations-tool-name.js";

 export { DEFAULT_SANDBOX_BROWSER_NETWORK } from "./browser-network.js";

@@ -46,7 +47,7 @@ export const DEFAULT_TOOL_DENY = [
  "computer",
  "mobile_ui",
  "nodes",
-  "cron",
+  AUTOMATIONS_TOOL_NAME,
  "gateway",
  ...CHANNEL_IDS,
 ] as const;
@@ -62,6 +62,7 @@ import type {
  ProviderSystemPromptSectionId,
 } from "./system-prompt-contribution.js";
 import type { PromptMode, SilentReplyPromptMode } from "./system-prompt.types.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";
 import {
  buildWatchedSessionsPromptLines,
  type PreparedWatchedSessionsPrompt,
@@ -863,7 +864,8 @@ export function buildAgentSystemPrompt(params: {
      "Own visible shell. Use for long/interactive jobs user should watch. exec for quiet work",
    canvas: "Present/eval/snapshot Canvas",
    nodes: "Paired node status/control/media",
-    cron: "Schedule/wake. Reminder text must read as reminder when fired; mention reminder for delayed gaps; include useful recent context.",
+    [AUTOMATIONS_TOOL_NAME]:
+      "Schedule/wake. Reminder text must read as reminder when fired; mention reminder for delayed gaps; include useful recent context.",
    message: "Message/channel actions",
    conversations_list: "List exact external conversation addresses",
    conversations_send: "Send directly to an external conversation",
@@ -905,7 +907,7 @@ export function buildAgentSystemPrompt(params: {
    "terminal",
    "canvas",
    "nodes",
-    "cron",
+    AUTOMATIONS_TOOL_NAME,
    "message",
    "conversations_list",
    "conversations_send",
@@ -16,7 +16,7 @@ vi.mock("../tools/computer-tool.js", () => ({
 }));

 vi.mock("../tools/cron-tool.js", () => ({
-  createCronTool: () => stubTool("cron"),
+  createCronTool: () => stubTool("automations"),
 }));

 vi.mock("../tools/gateway-tool.js", () => ({
@@ -25,7 +25,7 @@ function stubActionTool(name: string, actions: string[]) {
 const coreTools = [
  stubActionTool("canvas", ["create", "read"]),
  stubActionTool("nodes", ["list", "invoke"]),
-  stubActionTool("cron", ["schedule", "cancel"]),
+  stubActionTool("automations", ["schedule", "cancel"]),
  stubActionTool("message", ["send", "reply"]),
  stubTool("heartbeat_respond"),
  stubActionTool("gateway", ["config.get", "config.schema.lookup"]),
@@ -63,7 +63,7 @@ describe("tool-catalog", () => {
      "screen",
      "dashboard",
      "terminal",
-      "cron",
+      "automations",
      "get_goal",
      "create_goal",
      "update_goal",
@@ -23,6 +23,7 @@ import {
  DISMISS_TASK_TOOL_DISPLAY_SUMMARY,
  UPDATE_PLAN_TOOL_DISPLAY_SUMMARY,
 } from "./tool-description-presets.js";
+import { AUTOMATIONS_TOOL_NAME } from "./tools/automations-tool-name.js";

 /** Built-in tool profile ids exposed in config and UI. */
 export type ToolProfileId = "minimal" | "coding" | "messaging" | "full";
@@ -341,8 +342,8 @@ const CORE_TOOL_DEFINITIONS: CoreToolDefinition[] = [
    includeInOpenClawGroup: true,
  },
  {
-    id: "cron",
-    label: "cron",
+    id: AUTOMATIONS_TOOL_NAME,
+    label: AUTOMATIONS_TOOL_NAME,
    description: CRON_TOOL_DISPLAY_SUMMARY,
    sectionId: "automation",
    profiles: ["coding"],
@@ -1,4 +1,8 @@
 import { normalizeLowercaseStringOrEmpty } from "@openclaw/normalization-core/string-coerce";
+import {
+  AUTOMATIONS_TOOL_NAME,
+  LEGACY_AUTOMATIONS_TOOL_NAMES,
+} from "./tools/automations-tool-name.js";

 const MUTATING_TOOL_NAMES = new Set([
  "write",
@@ -11,7 +15,9 @@ const MUTATING_TOOL_NAMES = new Set([
  "sessions",
  "sessions_spawn",
  "sessions_send",
-  "cron",
+  AUTOMATIONS_TOOL_NAME,
+  // Saved transcripts predate the rename; legacy names must stay classified.
+  ...LEGACY_AUTOMATIONS_TOOL_NAMES,
  "gateway",
  "canvas",
  "computer",
@@ -267,7 +267,10 @@ describe("tool mutation helpers", () => {
        plan: [{ step: "Inspect", status: "in_progress" }],
      }),
    ).toBe(true);
+    expect(isReplaySafeToolCall("automations", { action: "status" })).toBe(true);
+    // Legacy transcript entries predate the rename and must stay classified.
    expect(isReplaySafeToolCall("cron", { action: "status" })).toBe(true);
+    expect(isReplaySafeToolCall("cron", { action: "add" })).toBe(false);
    expect(isReplaySafeToolCall("gateway", { action: "config.get" })).toBe(true);
    expect(isReplaySafeToolCall("gateway", { action: "config.schema.lookup" })).toBe(true);
    expect(isReplaySafeToolCall("gateway", { action: "config.patch" })).toBe(false);
@@ -9,6 +9,7 @@ import {
  normalizeOptionalLowercaseString,
 } from "@openclaw/normalization-core/string-coerce";
 import { isLikelyMutatingToolName } from "./tool-mutation-names.js";
+import { isAutomationsToolName } from "./tools/automations-tool-name.js";

 export { isLikelyMutatingToolName };

@@ -373,7 +374,7 @@ export function isMutatingToolCall(toolName: string, args: unknown): boolean {
    case "nodes":
      return action == null || !NODES_REPLAY_SAFE_ACTIONS.has(action);
    default: {
-      if (normalized === "cron" || normalized === "canvas") {
+      if (isAutomationsToolName(normalized) || normalized === "canvas") {
        return action == null || !READ_ONLY_ACTIONS.has(action);
      }
      if (normalized.endsWith("_actions")) {
@@ -424,7 +425,7 @@ export function isReplaySafeToolCall(toolName: string, args: unknown): boolean {
    case "nodes":
      return action != null && NODES_REPLAY_SAFE_ACTIONS.has(action);
    default: {
-      if (normalized === "cron" || normalized === "canvas") {
+      if (isAutomationsToolName(normalized) || normalized === "canvas") {
        return action != null && READ_ONLY_ACTIONS.has(action);
      }
      return false;
@@ -19,6 +19,8 @@ type ToolProfilePolicy = {
 const TOOL_NAME_ALIASES: Record<string, string> = {
  bash: "exec",
  "apply-patch": "apply_patch",
+  // Permanent scheduler-tool alias (owner decision, RFC 0026), like bash -> exec.
+  cron: "automations",
 };

 const TOOL_ALLOWLIST_INTERSECTION = Symbol.for("openclaw.toolAllowlistIntersection");
@@ -34,7 +34,7 @@ describe("tool-policy", () => {
  it("resolves known profiles and ignores unknown ones", () => {
    const coding = resolveToolProfilePolicy("coding");
    expect(coding?.allow).toContain("read");
-    expect(coding?.allow).toContain("cron");
+    expect(coding?.allow).toContain("automations");
    expect(coding?.allow).not.toContain("gateway");
    expect(resolveToolProfilePolicy("nope")).toBeUndefined();
  });
@@ -52,6 +52,9 @@ describe("tool-policy", () => {
    expect(normalizeToolName(" BASH ")).toBe("exec");
    expect(normalizeToolName("apply-patch")).toBe("apply_patch");
    expect(normalizeToolName("READ")).toBe("read");
+    // Pre-rename scheduler tool name from persisted config (RFC 0026).
+    expect(normalizeToolName("cron")).toBe("automations");
+    expect(normalizeToolName("automations")).toBe("automations");
  });

  it("collects explicit allowlist entries", () => {
@@ -180,6 +183,16 @@ describe("resolveSandboxToolPolicyForAgent", () => {
  });
 });

+describe("isToolAllowedByPolicyName — legacy scheduler tool name (RFC 0026)", () => {
+  it("allows the renamed tool through persisted legacy allow lists", () => {
+    expect(isToolAllowedByPolicyName("automations", { allow: ["cron"] })).toBe(true);
+  });
+
+  it("denies the renamed tool through persisted legacy deny lists", () => {
+    expect(isToolAllowedByPolicyName("automations", { deny: ["cron"] })).toBe(false);
+  });
+});
+
 describe("isToolAllowedByPolicyName — apply_patch / write deny decoupling (#76749)", () => {
  it("does not deny apply_patch when write is denied", () => {
    expect(isToolAllowedByPolicyName("apply_patch", { deny: ["write"] })).toBe(true);
@@ -102,7 +102,7 @@ const QUERY_EXPANSIONS: ReadonlyArray<{ terms: readonly string[]; add: readonly
  },
  {
    terms: ["remind", "reminder", "later", "tomorrow", "daily", "weekly", "recurring"],
-    add: ["schedule", "cron", "reminder"],
+    add: ["schedule", "automations", "cron", "reminder"],
  },
  { terms: ["say", "tell", "reply", "respond", "answer"], add: ["message", "send"] },
  { terms: ["picture", "photo", "meme", "screenshot"], add: ["image"] },
@@ -0,0 +1,22 @@
+/**
+ * Canonical identity of the scheduler agent tool. Single source of truth for
+ * every name-keyed consumer (policy lists, factory descriptors, runtime
+ * observers, prompts); never spell the tool name as a string literal.
+ */
+export const AUTOMATIONS_TOOL_NAME = "automations";
+
+/**
+ * "cron" is a permanently accepted alias for the scheduler tool in persisted
+ * allow/deny config, old transcripts, and inbound calls (owner decision,
+ * RFC 0026; same contract as bash -> exec). Not migration debt: no doctor
+ * rewrite, no removal window.
+ */
+export const LEGACY_AUTOMATIONS_TOOL_NAMES = ["cron"] as const;
+
+/** True when a tool name refers to the scheduler tool, including legacy names. */
+export function isAutomationsToolName(name: string): boolean {
+  return (
+    name === AUTOMATIONS_TOOL_NAME ||
+    (LEGACY_AUTOMATIONS_TOOL_NAMES as readonly string[]).includes(name)
+  );
+}
@@ -24,10 +24,11 @@ describe("cron tool creator cap", () => {
    capCronJobToolsAllowOnCreate(triggerJob, ["read", "cron"]);
    capCronJobToolsAllowOnCreate(plainJob, ["read", "cron"]);

+    // Legacy "cron" creator allowlists normalize to the canonical tool id.
    expect(triggerJob.payload).toEqual({
      kind: "systemEvent",
      text: "wake",
-      toolsAllow: ["read", "cron"],
+      toolsAllow: ["read", "automations"],
      toolsAllowIsDefault: true,
    });
    expect(plainJob.payload).toEqual({ kind: "systemEvent", text: "wake" });
@@ -103,7 +104,7 @@ describe("cron tool creator cap", () => {
      payload: {
        kind: "agentTurn",
        message: "updated",
-        toolsAllow: ["read", "cron"],
+        toolsAllow: ["read", "automations"],
        toolsAllowIsDefault: true,
      },
    });
@@ -1414,7 +1414,7 @@ describe("cron tool", () => {
    const params = expectSingleGatewayCallMethod("cron.add") as
      | { payload?: { toolsAllow?: string[] } }
      | undefined;
-    expect(params?.payload?.toolsAllow).toEqual(["read", "cron"]);
+    expect(params?.payload?.toolsAllow).toEqual(["read", "automations"]);
  });

  it("caps trigger-script systemEvent adds to the creator tool surface", async () => {
@@ -1437,7 +1437,7 @@ describe("cron tool", () => {
    const params = expectSingleGatewayCallMethod("cron.add") as
      | { payload?: { toolsAllow?: string[] } }
      | undefined;
-    expect(params?.payload?.toolsAllow).toEqual(["read", "cron"]);
+    expect(params?.payload?.toolsAllow).toEqual(["read", "automations"]);
  });

  it("infers systemEvent for implicit text payloads with toolsAllow", async () => {
@@ -1492,7 +1492,7 @@ describe("cron tool", () => {
          trigger: { script: "return { fire: false }" },
          payload: {
            kind: "systemEvent",
-            toolsAllow: ["read", "cron"],
+            toolsAllow: ["read", "automations"],
            toolsAllowIsDefault: true,
          },
        },
@@ -1576,7 +1576,7 @@ describe("cron tool", () => {
    expect(params?.payload?.toolsAllow).toEqual([
      "active_memory_search",
      "active_memory_store",
-      "cron",
+      "automations",
    ]);
  });

@@ -2973,7 +2973,7 @@ describe("cron tool", () => {
      | undefined;
    expect(params?.patch?.payload).toEqual({
      kind: "agentTurn",
-      toolsAllow: ["read", "cron"],
+      toolsAllow: ["read", "automations"],
      toolsAllowIsDefault: true,
    });
  });
@@ -3188,7 +3188,7 @@ describe("cron tool", () => {
          payload: {
            kind: "agentTurn",
            message: "run later",
-            toolsAllow: ["read", "cron"],
+            toolsAllow: ["read", "automations"],
            toolsAllowIsDefault: true,
          },
        },
@@ -20,6 +20,7 @@ import { resolveSessionAgentId } from "../agent-scope.js";
 import { CRON_TOOL_DISPLAY_SUMMARY } from "../tool-description-presets.js";
 import { normalizeToolName } from "../tool-policy.js";
 import { setToolTerminalPresentation } from "../tool-terminal-presentation.js";
+import { AUTOMATIONS_TOOL_NAME } from "./automations-tool-name.js";
 import {
  type AnyAgentTool,
  jsonResult,
@@ -284,8 +285,8 @@ function isOlderGatewayWithoutCompactCronList(error: unknown): boolean {
 export function createCronTool(opts?: CronToolOptions, deps?: CronToolDeps): AnyAgentTool {
  const callGateway = deps?.callGatewayTool ?? callGatewayTool;
  const tool: AnyAgentTool = {
-    label: "Cron",
-    name: "cron",
+    label: "Automations",
+    name: AUTOMATIONS_TOOL_NAME,
    displaySummary: CRON_TOOL_DISPLAY_SUMMARY,
    description: `Gateway scheduler: reminders, delayed self-wakeups, loops, recurring work, event watchers. Never exec sleep/poll as timer.

@@ -99,7 +99,7 @@ describe("gateway CLI backend live probe helpers", () => {
    activateLoopbackRuntime(port);
    try {
      await expect(verifyCliCronMcpLoopbackPreflight(preflightParams())).rejects.toThrow(
-        "mcp loopback tools/list did not expose cron",
+        "mcp loopback tools/list did not expose automations",
      );
      expect(methods).toEqual(["initialize", "notifications/initialized", "tools/list"]);
    } finally {
@@ -3,6 +3,7 @@
 import { randomUUID } from "node:crypto";
 import { normalizeLowercaseStringOrEmpty } from "@openclaw/normalization-core/string-coerce";
 import { renderCatFacePngBase64 } from "../../test/helpers/live-image-probe.js";
+import { AUTOMATIONS_TOOL_NAME } from "../agents/tools/automations-tool-name.js";
 import { isTruthyEnvValue } from "../infra/env.js";
 import { readResponseWithLimit } from "../infra/http-body.js";
 import { parseStrictPositiveInteger } from "../infra/parse-finite-number.js";
@@ -298,10 +299,10 @@ export async function verifyCliCronMcpLoopbackPreflight(params: {
    .filter(Boolean);
  logCliCronProbe("loopback-preflight:tools", {
    toolCount: toolNames.length,
-    cronVisible: toolNames.includes("cron"),
+    cronVisible: toolNames.includes(AUTOMATIONS_TOOL_NAME),
  });
-  if (!toolNames.includes("cron")) {
-    throw new Error("mcp loopback tools/list did not expose cron");
+  if (!toolNames.includes(AUTOMATIONS_TOOL_NAME)) {
+    throw new Error(`mcp loopback tools/list did not expose ${AUTOMATIONS_TOOL_NAME}`);
  }

  const toolCall = await callLoopbackJsonRpc({
@@ -314,7 +315,7 @@ export async function verifyCliCronMcpLoopbackPreflight(params: {
      id: "cron-add",
      method: "tools/call",
      params: {
-        name: "cron",
+        name: AUTOMATIONS_TOOL_NAME,
        arguments: JSON.parse(cronProbe.argsJson) as Record<string, unknown>,
      },
    },
@@ -67,8 +67,10 @@ describe("live-agent-probes", () => {
    expect(claudeRetryPrompt).toContain(
      "Preserve job.sessionTarget and job.sessionKey exactly as provided.",
    );
-    expect(claudeRetryPrompt).toContain("search/load MCP tools for `openclaw cron` or `cron`");
-    expect(claudeRetryPrompt).toContain("mcp__openclaw__cron");
+    expect(claudeRetryPrompt).toContain(
+      "search/load MCP tools for `openclaw automations` or `automations`",
+    );
+    expect(claudeRetryPrompt).toContain("mcp__openclaw__automations");
    expect(claudeRetryPrompt).toContain("Do not use Claude native `CronCreate`");
    expect(claudeRetryPrompt).not.toContain("openclaw-tools");
    expect(
@@ -86,7 +88,7 @@ describe("live-agent-probes", () => {
        attempt: 1,
        exactReply: spec.name,
      }),
-    ).toContain("previous OpenClaw cron MCP tool call was cancelled");
+    ).toContain("previous OpenClaw automations MCP tool call was cancelled");
    const args = JSON.parse(spec.argsJson) as {
      job?: {
        sessionTarget?: string;
@@ -113,9 +113,9 @@ export function buildLiveCronProbeMessage(params: {
  const claudeLike = isClaudeLikeLiveAgent(params.agent);
  if (params.attempt === 0) {
    return (
-      "Use the OpenClaw MCP cron tool from server `openclaw`. " +
-      "If it is not already visible, search/load MCP tools for `openclaw cron` or `cron`, " +
-      "then call the matching OpenClaw MCP tool; Claude-style names may appear as `mcp__openclaw__cron`. " +
+      "Use the OpenClaw MCP automations tool from server `openclaw`. " +
+      "If it is not already visible, search/load MCP tools for `openclaw automations` or `automations`, " +
+      "then call the matching OpenClaw MCP tool; Claude-style names may appear as `mcp__openclaw__automations`. " +
      "Do not use Claude native `CronCreate`, `CronList`, or `CronDelete`; those are not OpenClaw proof. " +
      `Call it with JSON arguments ${params.argsJson}. ` +
      "Preserve the JSON exactly, including job.sessionTarget and job.sessionKey; do not omit, rename, or flatten those fields. " +
@@ -125,9 +125,9 @@ export function buildLiveCronProbeMessage(params: {
  }
  if (claudeLike) {
    return (
-      "Retry the OpenClaw MCP cron tool from server `openclaw` now. " +
-      "If it is not already visible, search/load MCP tools for `openclaw cron` or `cron`, " +
-      "then call the matching OpenClaw MCP tool; Claude-style names may appear as `mcp__openclaw__cron`. " +
+      "Retry the OpenClaw MCP automations tool from server `openclaw` now. " +
+      "If it is not already visible, search/load MCP tools for `openclaw automations` or `automations`, " +
+      "then call the matching OpenClaw MCP tool; Claude-style names may appear as `mcp__openclaw__automations`. " +
      "Do not use Claude native `CronCreate`, `CronList`, or `CronDelete`; those are not OpenClaw proof. " +
      `Use these exact JSON arguments: ${params.argsJson}. ` +
      "Preserve job.sessionTarget and job.sessionKey exactly as provided. " +
@@ -138,9 +138,9 @@ export function buildLiveCronProbeMessage(params: {
    );
  }
  return (
-    "Your previous OpenClaw cron MCP tool call was cancelled before the job was created. " +
-    "Retry the OpenClaw MCP cron tool from server `openclaw` now. " +
-    "If the harness shows Claude-style MCP names, use `mcp__openclaw__cron`. " +
+    "Your previous OpenClaw automations MCP tool call was cancelled before the job was created. " +
+    "Retry the OpenClaw MCP automations tool from server `openclaw` now. " +
+    "If the harness shows Claude-style MCP names, use `mcp__openclaw__automations`. " +
    `Use these exact JSON arguments: ${params.argsJson}. ` +
    "Preserve job.sessionTarget and job.sessionKey exactly as provided. " +
    `If the cron job is created, reply exactly: ${params.exactReply}. ` +
@@ -9,6 +9,7 @@ import {
  resolveToolExecutionErrorKind,
  resolveToolResultFailureKind,
 } from "../agents/tool-result-error.js";
+import { isAutomationsToolName } from "../agents/tools/automations-tool-name.js";
 import type { McpLoopbackToolCallOutcome } from "./mcp-http.loopback-runtime.js";
 import {
  MCP_LOOPBACK_SERVER_NAME,
@@ -97,7 +98,17 @@ export async function handleMcpJsonRpc(params: {
    case "tools/list":
      return jsonRpcResult(id, { tools: params.toolSchema });
    case "tools/call": {
-      const toolName = typeof methodParams?.name === "string" ? methodParams.name.trim() : "";
+      const requestedToolName =
+        typeof methodParams?.name === "string" ? methodParams.name.trim() : "";
+      // "cron" is a permanently accepted inbound alias for the scheduler tool
+      // (owner decision, RFC 0026; same contract as bash -> exec). Resolve it to
+      // the published canonical tool without re-advertising it in tools/list.
+      const toolName =
+        !params.toolSchema.some((tool) => tool.name === requestedToolName) &&
+        isAutomationsToolName(requestedToolName)
+          ? (params.toolSchema.find((tool) => isAutomationsToolName(tool.name))?.name ??
+            requestedToolName)
+          : requestedToolName;
      const rawToolArgs = methodParams?.arguments;
      if (rawToolArgs !== undefined && !isRecord(rawToolArgs)) {
        return jsonRpcError(id, -32602, "Invalid params: tools/call arguments must be an object");
@@ -2834,6 +2834,44 @@ describe("mcp loopback server", () => {
    expect(signal).toBeInstanceOf(AbortSignal);
  });

+  it("resolves legacy cron tools/call names to the renamed automations tool", async () => {
+    const execute = vi.fn<MockGatewayTool["execute"]>(async () => ({
+      content: [{ type: "text", text: "SCHEDULED" }],
+    }));
+    const tool = makeMockTool({ name: "automations", description: "manage schedules", execute });
+
+    const payload = await handleMcpJsonRpc({
+      message: {
+        jsonrpc: "2.0",
+        id: 1,
+        method: "tools/call",
+        params: { name: "cron", arguments: {} },
+      },
+      tools: [tool as unknown as AnyAgentTool],
+      toolSchema: buildMockMcpToolSchema([tool]),
+    });
+
+    expectMcpResultText(payload as McpToolResultPayload, "SCHEDULED", false);
+    expect(execute).toHaveBeenCalledTimes(1);
+  });
+
+  it("still rejects legacy cron calls when the automations tool is not exposed", async () => {
+    const tool = makeMessageTool();
+
+    const payload = await handleMcpJsonRpc({
+      message: {
+        jsonrpc: "2.0",
+        id: 1,
+        method: "tools/call",
+        params: { name: "cron", arguments: {} },
+      },
+      tools: [tool as unknown as AnyAgentTool],
+      toolSchema: buildMockMcpToolSchema([tool]),
+    });
+
+    expectMcpResultText(payload as McpToolResultPayload, "Tool not available: cron", true);
+  });
+
  it("preserves request-disconnect evidence without classifying a tool failure", async () => {
    const controller = new AbortController();
    controller.abort();
@@ -8,6 +8,7 @@ import {
 } from "node:http";
 import { isRecord } from "@openclaw/normalization-core/record-coerce";
 import { resolveToolLoopDetectionConfig } from "../agents/tool-loop-detection-config.js";
+import { isAutomationsToolName } from "../agents/tools/automations-tool-name.js";
 import { getRuntimeConfig } from "../config/io.js";
 import { resolveSessionEntryAccessTarget } from "../config/sessions/session-accessor.js";
 import { isTruthyEnvValue } from "../infra/env.js";
@@ -340,7 +341,7 @@ async function startMcpLoopbackServer(port = 0): Promise<{
          inboundEventKind: requestContext.inboundEventKind,
          senderIsOwner: requestContext.senderIsOwner,
          toolCount: scopedTools.toolSchema.length,
-          cronVisible: scopedTools.toolSchema.some((tool) => tool.name === "cron"),
+          cronVisible: scopedTools.toolSchema.some((tool) => isAutomationsToolName(tool.name)),
        });
        const responses: object[] = [];
        for (const [messageIndex, message] of messages.entries()) {
@@ -78,7 +78,7 @@ const hoisted = vi.hoisted(() => {
    createOpenClawToolsMock: vi.fn((_args: CreateOpenClawToolsArg) => [
      makeTool("read"),
      makeTool("sessions_spawn"),
-      makeTool("cron"),
+      makeTool("automations"),
      makeTool("gateway"),
      makeTool("nodes"),
    ]),
@@ -144,7 +144,7 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {

    expect(result.tools.map((tool) => tool.name)).toEqual([
      "sessions_spawn",
-      "cron",
+      "automations",
      "gateway",
      "nodes",
    ]);
@@ -255,7 +255,7 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {
  it("does not fall back when policy removes a mediated coding tool", () => {
    hoisted.createOpenClawToolsMock.mockReturnValueOnce([
      hoisted.makeTool("write"),
-      hoisted.makeTool("cron"),
+      hoisted.makeTool("automations"),
    ]);

    const result = resolveGatewayScopedTools({
@@ -266,7 +266,7 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {
      excludeToolNames: ["read", "edit", "apply_patch", "exec", "process"],
    });

-    expect(result.tools.map((tool) => tool.name)).toEqual(["cron"]);
+    expect(result.tools.map((tool) => tool.name)).toEqual(["automations"]);
  });

  it("keeps owner-only core tools visible only for owner loopback callers", () => {
@@ -290,14 +290,14 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {
    expect(ownerResult.tools.map((tool) => tool.name)).toEqual([
      "read",
      "sessions_spawn",
-      "cron",
+      "automations",
      "gateway",
      "nodes",
    ]);
    expect(nonOwnerResult.tools.map((tool) => tool.name)).toEqual(["read", "sessions_spawn"]);
    const args = readCreateToolsArgs(1);
    expect(args.pluginToolDenylist).toEqual([
-      "cron",
+      "automations",
      "gateway",
      "sessions",
      "screen",
@@ -311,7 +311,7 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {
      "openclaw",
    ]);
    expect(args.inheritedToolDenylist).toEqual([
-      "cron",
+      "automations",
      "gateway",
      "sessions",
      "screen",
@@ -804,7 +804,7 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {

  it("does not inherit node-only exec as a generic child or cron capability", () => {
    const result = resolveGatewayScopedTools({
-      cfg: { tools: { allow: ["exec", "sessions_spawn", "cron"] } } as OpenClawConfig,
+      cfg: { tools: { allow: ["exec", "sessions_spawn", "automations"] } } as OpenClawConfig,
      sessionKey: "agent:main:direct:test",
      surface: "loopback",
      senderIsOwner: true,
@@ -820,7 +820,7 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {
    const result = resolveGatewayScopedTools({
      cfg: {
        agents: { defaults: { sandbox: { mode: "all" } } },
-        tools: { sandbox: { tools: { deny: ["cron"] } } },
+        tools: { sandbox: { tools: { deny: ["automations"] } } },
      } as OpenClawConfig,
      sessionKey: "agent:main:direct:test",
      surface: "loopback",
@@ -829,36 +829,36 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {
    expect(result.tools.map((tool) => tool.name)).toEqual(["read", "sessions_spawn"]);
    const args = readCreateToolsArgs();
    expect(args.sandboxed).toBe(true);
-    expect(args.pluginToolDenylist).toEqual(["cron"]);
-    expect(args.inheritedToolDenylist).toEqual(["cron"]);
+    expect(args.pluginToolDenylist).toEqual(["automations"]);
+    expect(args.inheritedToolDenylist).toEqual(["automations"]);
  });

  it("passes final filtered tool surface to gateway cron jobs", () => {
    hoisted.createOpenClawToolsMock.mockReturnValueOnce([
      hoisted.makeTool("read"),
-      hoisted.makeTool("cron"),
+      hoisted.makeTool("automations"),
      hoisted.makeTool("exec"),
    ]);

    const result = resolveGatewayScopedTools({
      cfg: {
-        tools: { allow: ["read", "cron"] },
+        tools: { allow: ["read", "automations"] },
      } as OpenClawConfig,
      sessionKey: "agent:main:direct:test",
      surface: "loopback",
    });

-    expect(result.tools.map((tool) => tool.name)).toEqual(["read", "cron"]);
+    expect(result.tools.map((tool) => tool.name)).toEqual(["read", "automations"]);
    expect(readCreateToolsArgs().cronCreatorToolAllowlist).toEqual([
      { name: "read" },
-      { name: "cron" },
+      { name: "automations" },
    ]);
  });

  it("passes unrestricted gateway tool surfaces to cron jobs", () => {
    hoisted.createOpenClawToolsMock.mockReturnValueOnce([
      hoisted.makeTool("read"),
-      hoisted.makeTool("cron"),
+      hoisted.makeTool("automations"),
      hoisted.makeTool("exec"),
    ]);

@@ -869,10 +869,10 @@ describe("resolveGatewayScopedTools excludeToolNames", () => {
      senderIsOwner: true,
    });

-    expect(result.tools.map((tool) => tool.name)).toEqual(["read", "cron", "exec"]);
+    expect(result.tools.map((tool) => tool.name)).toEqual(["read", "automations", "exec"]);
    expect(readCreateToolsArgs().cronCreatorToolAllowlist).toEqual([
      { name: "read" },
-      { name: "cron" },
+      { name: "automations" },
      { name: "exec" },
    ]);
  });
@@ -193,7 +193,14 @@ export function resolveGatewayScopedTools(params: {
  const gatewayToolsCfg = params.cfg.gateway?.tools;
  const defaultGatewayDeny =
    surface === "http"
-      ? DEFAULT_GATEWAY_HTTP_TOOL_DENY.filter((name) => !gatewayToolsCfg?.allow?.includes(name))
+      ? DEFAULT_GATEWAY_HTTP_TOOL_DENY.filter(
+          // Config allow entries may use legacy tool names (e.g. "cron");
+          // normalize both sides so they still lift the matching default deny.
+          (name) =>
+            !gatewayToolsCfg?.allow?.some(
+              (allowed) => normalizeToolName(allowed) === normalizeToolName(name),
+            ),
+        )
      : [];
  const ownerOnlyGatewayDeny =
    params.senderIsOwner === false || (surface === "http" && params.senderIsOwner !== true)
@@ -60,9 +60,9 @@ vi.mock("../plugins/tools.js", () => ({
 vi.mock("../agents/openclaw-tools.js", () => {
  const tools = [
    {
-      name: "cron",
+      name: "automations",
      parameters: { type: "object", properties: { action: { type: "string" } } },
-      execute: async () => ({ ok: true, via: "cron" }),
+      execute: async () => ({ ok: true, via: "automations" }),
    },
    {
      name: "gateway",
@@ -141,9 +141,9 @@ vi.mock("../agents/openclaw-tools.js", () => {
      },
    },
    {
-      name: "cron",
+      name: "automations",
      parameters: { type: "object", properties: {} },
-      execute: async () => ({ ok: true, result: "cron" }),
+      execute: async () => ({ ok: true, result: "automations" }),
    },
    {
      name: "exec",
@@ -797,7 +797,7 @@ describe("POST /tools/invoke", () => {

    const body = await expectOkInvokeResponse(res);
    expect(body.result?.inheritedToolDenylist).toEqual(
-      expect.arrayContaining(["cron", "gateway", "nodes"]),
+      expect.arrayContaining(["automations", "gateway", "nodes"]),
    );
  });

@@ -1227,7 +1227,8 @@ describe("tools.invoke Gateway RPC", () => {

      expect(call?.[0], tool).toBe(true);
      expect(call?.[1]?.ok, tool).toBe(false);
-      expect(call?.[1]?.toolName, tool).toBe(tool);
+      // Legacy "cron" requests canonicalize before dispatch and report the canonical id.
+      expect(call?.[1]?.toolName, tool).toBe(tool === "cron" ? "automations" : tool);
      const error = call?.[1]?.error as { code?: string; message?: string } | undefined;
      expect(error?.code, tool).toBe("not_found");
    }
@@ -8,6 +8,10 @@ import { runBeforeToolCallHook } from "../agents/agent-tools.before-tool-call.js
 import { resolveToolLoopDetectionConfig } from "../agents/agent-tools.js";
 import { getChannelAgentToolMeta } from "../agents/channel-tools.js";
 import { isKnownCoreToolId } from "../agents/tool-catalog.js";
+import {
+  AUTOMATIONS_TOOL_NAME,
+  isAutomationsToolName,
+} from "../agents/tools/automations-tool-name.js";
 import { ToolInputError, type AnyAgentTool } from "../agents/tools/common.js";
 import {
  normalizeConversationReadInvocationOrigin,
@@ -174,7 +178,13 @@ export async function invokeGatewayTool(params: {
  const conversationReadOrigin = normalizeConversationReadInvocationOrigin(
    params.conversationReadOrigin,
  );
-  const toolName = normalizeOptionalString(params.input.name ?? params.input.tool) ?? "";
+  const requestedToolName = normalizeOptionalString(params.input.name ?? params.input.tool) ?? "";
+  // "cron" is a permanently accepted inbound alias for the scheduler tool
+  // (owner decision, RFC 0026; same contract as bash -> exec). Canonicalize
+  // before core-id checks and exact-name dispatch below.
+  const toolName = isAutomationsToolName(requestedToolName)
+    ? AUTOMATIONS_TOOL_NAME
+    : requestedToolName;
  if (!toolName) {
    return {
      ok: false,
@@ -28,7 +28,7 @@ describe("OpenClaw tools MCP server", () => {
    );

    const listed = await handlers.listTools();
-    expect(listed.tools.map((tool) => tool.name)).toContain("cron");
+    expect(listed.tools.map((tool) => tool.name)).toContain("automations");
  });

  it("requires the managed bridge to pass a real agent session key", () => {
@@ -6,6 +6,7 @@
 */
 import { pathToFileURL } from "node:url";
 import { Server } from "@modelcontextprotocol/sdk/server/index.js";
+import { AUTOMATIONS_TOOL_NAME } from "../agents/tools/automations-tool-name.js";
 import type { AnyAgentTool } from "../agents/tools/common.js";
 import { createCronTool } from "../agents/tools/cron-tool.js";
 import { createSystemAgentTool } from "../agents/tools/system-agent-tool.js";
@@ -57,7 +58,10 @@ export function resolveOpenClawToolsForMcp(
    if (!agentSessionKey) {
      throw new Error(`${OPENCLAW_TOOLS_MCP_AGENT_SESSION_KEY_ENV} is required`);
    }
-    return createCronTool({ agentSessionKey, creatorToolAllowlist: [{ name: "cron" }] });
+    return createCronTool({
+      agentSessionKey,
+      creatorToolAllowlist: [{ name: AUTOMATIONS_TOOL_NAME }],
+    });
  });
 }

@@ -8,6 +8,7 @@ import {
  wrapToolWithBeforeToolCallHook,
 } from "../agents/agent-tools.before-tool-call.js";
 import { BEFORE_TOOL_CALL_HOOK_CONTEXT } from "../agents/before-tool-call-metadata.js";
+import { isAutomationsToolName } from "../agents/tools/automations-tool-name.js";
 import type { AnyAgentTool } from "../agents/tools/common.js";
 import { formatErrorMessage } from "../infra/errors.js";
 import { coerceChatContentText } from "../shared/chat-content.js";
@@ -82,7 +83,14 @@ export function createPluginToolsMcpHandlers(tools: AnyAgentTool[]) {
      })),
    }),
    callTool: async (params: CallPluginToolParams, signal?: AbortSignal) => {
-      const entry = toolMap.get(params.name);
+      // "cron" is a permanently accepted inbound alias for the scheduler tool
+      // (owner decision, RFC 0026; same contract as bash -> exec). Resolve it to
+      // the published canonical tool without re-advertising it in listTools.
+      const entry =
+        toolMap.get(params.name) ??
+        (isAutomationsToolName(params.name)
+          ? Array.from(toolMap.entries()).find(([name]) => isAutomationsToolName(name))?.[1]
+          : undefined);
      if (!entry) {
        return {
          content: [{ type: "text", text: `Unknown tool: ${params.name}` }],
@@ -244,8 +244,8 @@ describe("security audit trust model findings", () => {
          );
          expect(finding?.severity).toBe("critical");
          expect(finding?.detail).toContain("channels.whatsapp.groupPolicy");
-          expect(finding?.detail).toContain("controlPlane=[cron]");
-          expect(finding?.detail).not.toContain("controlPlane=[gateway, cron]");
+          expect(finding?.detail).toContain("controlPlane=[automations]");
+          expect(finding?.detail).not.toContain("controlPlane=[automations, gateway]");
        },
      },
      {
@@ -278,7 +278,7 @@ describe("security audit trust model findings", () => {
            (entry) => entry.checkId === "security.exposure.open_groups_with_control_plane_tools",
          );
          expect(finding?.detail).toContain(
-            "agents.defaults (profile=messaging; controlPlane=[cron])",
+            "agents.defaults (profile=messaging; controlPlane=[automations])",
          );
        },
      },
@@ -1,5 +1,6 @@
 // Shared tool-risk constants.
 // Keep these centralized so gateway HTTP restrictions and security audits don't drift.
+import { AUTOMATIONS_TOOL_NAME } from "../agents/tools/automations-tool-name.js";

 /**
 * Tools denied via Gateway HTTP `POST /tools/invoke` by default.
@@ -32,7 +33,7 @@ export const DEFAULT_GATEWAY_HTTP_TOOL_DENY = [
  "conversations_send",
  "conversations_turn",
  // Persistent automation control plane — can create/update/remove scheduled runs
-  "cron",
+  AUTOMATIONS_TOOL_NAME,
  // Gateway config can expose secrets and host topology
  "gateway",
  // Node command relay can reach system.run on paired hosts
@@ -45,10 +46,10 @@ export const DEFAULT_GATEWAY_HTTP_TOOL_DENY = [
 ] as const;

 /**
- * Sensitive control-plane tools. `cron` can persist automation; `gateway`
+ * Sensitive control-plane tools. `automations` can persist scheduled runs; `gateway`
 * exposes configuration and schema details even though its agent actions are read-only.
 */
-export const GATEWAY_CONTROL_PLANE_TOOLS = ["cron", "gateway"] as const;
+export const GATEWAY_CONTROL_PLANE_TOOLS = [AUTOMATIONS_TOOL_NAME, "gateway"] as const;

 /**
 * Core tools that require sender owner identity on Gateway-scoped surfaces.
@@ -59,7 +59,7 @@ describe("resolveSkillDispatchTools", () => {

    const args = hoisted.createOpenClawToolsMock.mock.calls[0]?.[0];
    expect(tools.map((tool) => tool.name)).toEqual(["read", "cron"]);
-    expect(args?.cronCreatorToolAllowlist).toEqual([{ name: "read" }, { name: "cron" }]);
+    expect(args?.cronCreatorToolAllowlist).toEqual([{ name: "read" }, { name: "automations" }]);
    expect(args?.nativeChannelId).toBe("native-room-1");
  });

@@ -78,7 +78,7 @@ describe("resolveSkillDispatchTools", () => {
    expect(tools.map((tool) => tool.name)).toEqual(["read", "cron", "exec"]);
    expect(args?.cronCreatorToolAllowlist).toEqual([
      { name: "read" },
-      { name: "cron" },
+      { name: "automations" },
      { name: "exec" },
    ]);
  });
@@ -1,6 +1,7 @@
 /** In-memory spoken confirmation binding for high-impact Talk actions. */
 import { createHash, randomUUID } from "node:crypto";
 import { buildToolMutationState } from "../agents/tool-mutation.js";
+import { AUTOMATIONS_TOOL_NAME } from "../agents/tools/automations-tool-name.js";

 const CONFIRMATION_TTL_MS = 2 * 60_000;

@@ -72,7 +73,7 @@ function requiresHighImpactVoiceConfirmation(toolName: string, params: unknown):
      "computer",
      "mobile_ui",
      "canvas",
-      "cron",
+      AUTOMATIONS_TOOL_NAME,
      "process",
    ].includes(normalizedTool)
  ) {
@@ -1100,7 +1100,7 @@
          "required": ["action"],
          "type": "object"
        },
-        "name": "cron",
+        "name": "automations",
        "type": "function"
      },
      {
@@ -1096,7 +1096,7 @@
          "required": ["action"],
          "type": "object"
        },
-        "name": "cron",
+        "name": "automations",
        "type": "function"
      },
      {
@@ -1096,7 +1096,7 @@
          "required": ["action"],
          "type": "object"
        },
-        "name": "cron",
+        "name": "automations",
        "type": "function"
      },
      {
@@ -78,7 +78,7 @@
    "agents_list",
    "message",
    "sessions_spawn",
-    "cron",
+    "automations",
    "gateway",
    "nodes",
    "session_status",
@@ -221,20 +221,20 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
    "roughTokens": 0
  },
  "dynamicToolsJson": {
-    "chars": 61437,
-    "roughTokens": 15360
+    "chars": 61444,
+    "roughTokens": 15361
  },
  "openClawDeveloperInstructions": {
-    "chars": 3804,
-    "roughTokens": 951
+    "chars": 3811,
+    "roughTokens": 953
  },
  "totalTextOnly": {
-    "chars": 28187,
-    "roughTokens": 7047
+    "chars": 28194,
+    "roughTokens": 7049
  },
  "totalWithDynamicToolsJson": {
-    "chars": 89626,
-    "roughTokens": 22407
+    "chars": 89640,
+    "roughTokens": 22410
  },
  "userInputText": {
    "chars": 1300,
@@ -421,7 +421,7 @@ Approval policy is currently never. Do not provide the `sandbox_permissions` for
 ````text
 You are a personal agent running inside OpenClaw. OpenClaw has dynamic tools for OpenClaw-owned messaging, cron, sessions, media, gateway, and nodes.

-Deferred searchable OpenClaw dynamic tools available: cron, gateway, nodes, session_status, sessions_history, sessions_list, sessions_search, sessions_send, subagents, tts, web_fetch, web_search. Use `tool_search` to load exact callable specs before use.
+Deferred searchable OpenClaw dynamic tools available: automations, gateway, nodes, session_status, sessions_history, sessions_list, sessions_search, sessions_send, subagents, tts, web_fetch, web_search. Use `tool_search` to load exact callable specs before use.

 Use Codex native `spawn_agent` for Codex subagents. `spawn_agent` and the other native collaboration tools may be deferred: when `spawn_agent` is not directly listed, load it with `tool_search` before spawning. Use OpenClaw `sessions_spawn` only for OpenClaw or ACP delegation, never as a substitute for `spawn_agent`.

@@ -526,7 +526,7 @@ Full JSON: `codex-dynamic-tools.discord-group.json`
  "agents_list",
  "message",
  "sessions_spawn",
-  "cron",
+  "automations",
  "gateway",
  "nodes",
  "session_status",
@@ -78,7 +78,7 @@
    "agents_list",
    "message",
    "sessions_spawn",
-    "cron",
+    "automations",
    "gateway",
    "nodes",
    "session_status",
@@ -221,20 +221,20 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
    "roughTokens": 0
  },
  "dynamicToolsJson": {
-    "chars": 61129,
-    "roughTokens": 15283
+    "chars": 61136,
+    "roughTokens": 15284
  },
  "openClawDeveloperInstructions": {
-    "chars": 2695,
-    "roughTokens": 674
+    "chars": 2702,
+    "roughTokens": 676
  },
  "totalTextOnly": {
-    "chars": 26707,
-    "roughTokens": 6677
+    "chars": 26714,
+    "roughTokens": 6679
  },
  "totalWithDynamicToolsJson": {
-    "chars": 87838,
-    "roughTokens": 21960
+    "chars": 87852,
+    "roughTokens": 21963
  },
  "userInputText": {
    "chars": 929,
@@ -421,7 +421,7 @@ Approval policy is currently never. Do not provide the `sandbox_permissions` for
 ````text
 You are a personal agent running inside OpenClaw. OpenClaw has dynamic tools for OpenClaw-owned messaging, cron, sessions, media, gateway, and nodes.

-Deferred searchable OpenClaw dynamic tools available: cron, gateway, nodes, session_status, sessions_history, sessions_list, sessions_search, sessions_send, subagents, tts, web_fetch, web_search. Use `tool_search` to load exact callable specs before use.
+Deferred searchable OpenClaw dynamic tools available: automations, gateway, nodes, session_status, sessions_history, sessions_list, sessions_search, sessions_send, subagents, tts, web_fetch, web_search. Use `tool_search` to load exact callable specs before use.

 Use Codex native `spawn_agent` for Codex subagents. `spawn_agent` and the other native collaboration tools may be deferred: when `spawn_agent` is not directly listed, load it with `tool_search` before spawning. Use OpenClaw `sessions_spawn` only for OpenClaw or ACP delegation, never as a substitute for `spawn_agent`.

@@ -520,7 +520,7 @@ Full JSON: `codex-dynamic-tools.telegram-direct.json`
  "agents_list",
  "message",
  "sessions_spawn",
-  "cron",
+  "automations",
  "gateway",
  "nodes",
  "session_status",
@@ -78,7 +78,7 @@
    "agents_list",
    "message",
    "sessions_spawn",
-    "cron",
+    "automations",
    "gateway",
    "nodes",
    "session_status",
@@ -222,20 +222,20 @@ This is the deterministic model-bound layer stack OpenClaw can snapshot for the
    "roughTokens": 0
  },
  "dynamicToolsJson": {
-    "chars": 62661,
-    "roughTokens": 15666
+    "chars": 62668,
+    "roughTokens": 15667
  },
  "openClawDeveloperInstructions": {
-    "chars": 2695,
-    "roughTokens": 674
+    "chars": 2702,
+    "roughTokens": 676
  },
  "totalTextOnly": {
-    "chars": 27136,
-    "roughTokens": 6784
+    "chars": 27143,
+    "roughTokens": 6786
  },
  "totalWithDynamicToolsJson": {
-    "chars": 89799,
-    "roughTokens": 22450
+    "chars": 89813,
+    "roughTokens": 22454
  },
  "userInputText": {
    "chars": 1284,
@@ -422,7 +422,7 @@ Approval policy is currently never. Do not provide the `sandbox_permissions` for
 ````text
 You are a personal agent running inside OpenClaw. OpenClaw has dynamic tools for OpenClaw-owned messaging, cron, sessions, media, gateway, and nodes.

-Deferred searchable OpenClaw dynamic tools available: cron, gateway, nodes, session_status, sessions_history, sessions_list, sessions_search, sessions_send, subagents, tts, web_fetch, web_search. Use `tool_search` to load exact callable specs before use.
+Deferred searchable OpenClaw dynamic tools available: automations, gateway, nodes, session_status, sessions_history, sessions_list, sessions_search, sessions_send, subagents, tts, web_fetch, web_search. Use `tool_search` to load exact callable specs before use.

 Use Codex native `spawn_agent` for Codex subagents. `spawn_agent` and the other native collaboration tools may be deferred: when `spawn_agent` is not directly listed, load it with `tool_search` before spawning. Use OpenClaw `sessions_spawn` only for OpenClaw or ACP delegation, never as a substitute for `spawn_agent`.

@@ -521,7 +521,7 @@ Full JSON: `codex-dynamic-tools.heartbeat-turn.json`
  "agents_list",
  "message",
  "sessions_spawn",
-  "cron",
+  "automations",
  "gateway",
  "nodes",
  "session_status",
@@ -59,7 +59,7 @@ const CODEX_YOLO_PERMISSION_INSTRUCTIONS = [
 ].join("\n");
 const HAPPY_PATH_TOOL_NAMES = new Set([
  "nodes",
-  "cron",
+  "automations",
  "message",
  "heartbeat_respond",
  "tts",