diff --git a/CHANGELOG.md b/CHANGELOG.md
index 36d4a793818..f9a33b6b012 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -23,6 +23,7 @@ Docs: https://docs.openclaw.ai
 
 ### Fixes
 
+- Security/audit: resolve configured model aliases before model-tier and small-parameter checks, so alias-based GPT-5/Codex configs no longer report false weak-model warnings. Fixes #74455. Thanks @blaspat.
 - Models/UI: hide unauthenticated providers from the default Web chat, `/models`, and model setup pickers while keeping explicit full-catalog browse paths through `view: "all"`, `/models <provider> all`, and `models list --all`. Fixes #74423. Thanks @guarismo and @SymbolStar.
 - Slack/prompts: rely on Slack `interactiveReplies` guidance instead of generic `inlineButtons` config hints so enabled Slack button directives are not contradicted. Fixes #46647. Thanks @jeremykoerber.
 - Slack/reactions: treat duplicate `already_reacted` responses as idempotent success so repeated agent reaction adds no longer surface as tool failures. Fixes #69005. Thanks @shipitsteven and @martingarramon.
diff --git a/src/security/audit-extra.summary.ts b/src/security/audit-extra.summary.ts
index edada8549e1..0011bd1900a 100644
--- a/src/security/audit-extra.summary.ts
+++ b/src/security/audit-extra.summary.ts
@@ -3,15 +3,12 @@ import { resolveSandboxToolPolicyForAgent } from "../agents/sandbox/tool-policy.
 import type { SandboxToolPolicy } from "../agents/sandbox/types.js";
 import { isToolAllowedByPolicies } from "../agents/tool-policy-match.js";
 import { resolveToolProfilePolicy } from "../agents/tool-policy.js";
-import {
-  resolveAgentModelFallbackValues,
-  resolveAgentModelPrimaryValue,
-} from "../config/model-input.js";
 import type { OpenClawConfig } from "../config/types.openclaw.js";
 import type { AgentToolsConfig } from "../config/types.tools.js";
 import { hasConfiguredInternalHooks } from "../hooks/configured.js";
 import { hasConfiguredWebSearchCredential } from "../plugins/web-search-credential-presence.js";
 import { inferParamBFromIdOrName } from "../shared/model-param-b.js";
+import { collectAuditModelRefs } from "./audit-model-refs.js";
 import { pickSandboxToolPolicy } from "./audit-tool-policy.js";
 
 export type SecurityAuditFinding = {
@@ -24,8 +21,6 @@ export type SecurityAuditFinding = {
 
 const SMALL_MODEL_PARAM_B_MAX = 300;
 
-type ModelRef = { id: string; source: string };
-
 function summarizeGroupPolicy(cfg: OpenClawConfig): {
   open: number;
   allowlist: number;
@@ -55,59 +50,6 @@ function summarizeGroupPolicy(cfg: OpenClawConfig): {
   return { open, allowlist, other };
 }
 
-function addModel(models: ModelRef[], raw: unknown, source: string) {
-  if (typeof raw !== "string") {
-    return;
-  }
-  const id = raw.trim();
-  if (!id) {
-    return;
-  }
-  models.push({ id, source });
-}
-
-function collectModels(cfg: OpenClawConfig): ModelRef[] {
-  const out: ModelRef[] = [];
-  addModel(
-    out,
-    resolveAgentModelPrimaryValue(cfg.agents?.defaults?.model),
-    "agents.defaults.model.primary",
-  );
-  for (const fallback of resolveAgentModelFallbackValues(cfg.agents?.defaults?.model)) {
-    addModel(out, fallback, "agents.defaults.model.fallbacks");
-  }
-  addModel(
-    out,
-    resolveAgentModelPrimaryValue(cfg.agents?.defaults?.imageModel),
-    "agents.defaults.imageModel.primary",
-  );
-  for (const fallback of resolveAgentModelFallbackValues(cfg.agents?.defaults?.imageModel)) {
-    addModel(out, fallback, "agents.defaults.imageModel.fallbacks");
-  }
-
-  const list = Array.isArray(cfg.agents?.list) ? cfg.agents?.list : [];
-  for (const agent of list ?? []) {
-    if (!agent || typeof agent !== "object") {
-      continue;
-    }
-    const id =
-      typeof (agent as { id?: unknown }).id === "string" ? (agent as { id: string }).id : "";
-    const model = (agent as { model?: unknown }).model;
-    if (typeof model === "string") {
-      addModel(out, model, `agents.list.${id}.model`);
-    } else if (model && typeof model === "object") {
-      addModel(out, (model as { primary?: unknown }).primary, `agents.list.${id}.model.primary`);
-      const fallbacks = (model as { fallbacks?: unknown }).fallbacks;
-      if (Array.isArray(fallbacks)) {
-        for (const fallback of fallbacks) {
-          addModel(out, fallback, `agents.list.${id}.model.fallbacks`);
-        }
-      }
-    }
-  }
-  return out;
-}
-
 function extractAgentIdFromSource(source: string): string | null {
   const match = source.match(/^agents\.list\.([^.]*)\./);
   return match?.[1] ?? null;
@@ -210,7 +152,9 @@ export function collectSmallModelRiskFindings(params: {
   env: NodeJS.ProcessEnv;
 }): SecurityAuditFinding[] {
   const findings: SecurityAuditFinding[] = [];
-  const models = collectModels(params.cfg).filter((entry) => !entry.source.includes("imageModel"));
+  const models = collectAuditModelRefs(params.cfg).filter(
+    (entry) => !entry.source.includes("imageModel"),
+  );
   if (models.length === 0) {
     return findings;
   }
diff --git a/src/security/audit-extra.sync.ts b/src/security/audit-extra.sync.ts
index b094e16c665..cfa7c9e7773 100644
--- a/src/security/audit-extra.sync.ts
+++ b/src/security/audit-extra.sync.ts
@@ -6,10 +6,6 @@ import { getBlockedBindReason } from "../agents/sandbox/validate-sandbox-securit
 import { isToolAllowedByPolicies } from "../agents/tool-policy-match.js";
 import { resolveToolProfilePolicy } from "../agents/tool-policy.js";
 import { formatCliCommand } from "../cli/command-format.js";
-import {
-  resolveAgentModelFallbackValues,
-  resolveAgentModelPrimaryValue,
-} from "../config/model-input.js";
 import type { OpenClawConfig } from "../config/types.openclaw.js";
 import type { AgentToolsConfig } from "../config/types.tools.js";
 import { resolveGatewayAuth } from "../gateway/auth.js";
@@ -23,6 +19,7 @@ import {
   normalizeOptionalString,
   normalizeStringifiedOptionalString,
 } from "../shared/string-coerce.js";
+import { collectAuditModelRefs } from "./audit-model-refs.js";
 import { pickSandboxToolPolicy } from "./audit-tool-policy.js";
 
 /**
@@ -59,61 +56,6 @@ function looksLikeEnvRef(value: string): boolean {
   return v.startsWith("${") && v.endsWith("}");
 }
 
-type ModelRef = { id: string; source: string };
-
-function addModel(models: ModelRef[], raw: unknown, source: string) {
-  if (typeof raw !== "string") {
-    return;
-  }
-  const id = raw.trim();
-  if (!id) {
-    return;
-  }
-  models.push({ id, source });
-}
-
-function collectModels(cfg: OpenClawConfig): ModelRef[] {
-  const out: ModelRef[] = [];
-  addModel(
-    out,
-    resolveAgentModelPrimaryValue(cfg.agents?.defaults?.model),
-    "agents.defaults.model.primary",
-  );
-  for (const fallback of resolveAgentModelFallbackValues(cfg.agents?.defaults?.model)) {
-    addModel(out, fallback, "agents.defaults.model.fallbacks");
-  }
-  addModel(
-    out,
-    resolveAgentModelPrimaryValue(cfg.agents?.defaults?.imageModel),
-    "agents.defaults.imageModel.primary",
-  );
-  for (const fallback of resolveAgentModelFallbackValues(cfg.agents?.defaults?.imageModel)) {
-    addModel(out, fallback, "agents.defaults.imageModel.fallbacks");
-  }
-
-  const list = Array.isArray(cfg.agents?.list) ? cfg.agents?.list : [];
-  for (const agent of list ?? []) {
-    if (!agent || typeof agent !== "object") {
-      continue;
-    }
-    const id =
-      typeof (agent as { id?: unknown }).id === "string" ? (agent as { id: string }).id : "";
-    const model = (agent as { model?: unknown }).model;
-    if (typeof model === "string") {
-      addModel(out, model, `agents.list.${id}.model`);
-    } else if (model && typeof model === "object") {
-      addModel(out, (model as { primary?: unknown }).primary, `agents.list.${id}.model.primary`);
-      const fallbacks = (model as { fallbacks?: unknown }).fallbacks;
-      if (Array.isArray(fallbacks)) {
-        for (const fallback of fallbacks) {
-          addModel(out, fallback, `agents.list.${id}.model.fallbacks`);
-        }
-      }
-    }
-  }
-  return out;
-}
-
 function isGatewayRemotelyExposed(cfg: OpenClawConfig): boolean {
   const bind = typeof cfg.gateway?.bind === "string" ? cfg.gateway.bind : "loopback";
   if (bind !== "loopback") {
@@ -986,7 +928,7 @@ export function collectMinimalProfileOverrideFindings(cfg: OpenClawConfig): Secu
 
 export function collectModelHygieneFindings(cfg: OpenClawConfig): SecurityAuditFinding[] {
   const findings: SecurityAuditFinding[] = [];
-  const models = collectModels(cfg);
+  const models = collectAuditModelRefs(cfg);
   if (models.length === 0) {
     return findings;
   }
diff --git a/src/security/audit-model-hygiene.test.ts b/src/security/audit-model-hygiene.test.ts
index 45f573c6601..6ceb7f31ddf 100644
--- a/src/security/audit-model-hygiene.test.ts
+++ b/src/security/audit-model-hygiene.test.ts
@@ -52,4 +52,24 @@ describe("security audit model hygiene findings", () => {
       }
     }
   });
+
+  it("resolves configured aliases before tier classification", () => {
+    const findings = collectModelHygieneFindings({
+      agents: {
+        defaults: {
+          model: {
+            primary: "gpt",
+            fallbacks: ["gpt-prev", "gpt-mini"],
+          },
+          models: {
+            "openai-codex/gpt-5.5": { alias: "gpt" },
+            "openai-codex/gpt-5.4": { alias: "gpt-prev" },
+            "openai/gpt-5-mini": { alias: "gpt-mini" },
+          },
+        },
+      },
+    } satisfies OpenClawConfig);
+
+    expect(findings.some((finding) => finding.checkId === "models.weak_tier")).toBe(false);
+  });
 });
diff --git a/src/security/audit-model-refs.ts b/src/security/audit-model-refs.ts
new file mode 100644
index 00000000000..cd20ce55e40
--- /dev/null
+++ b/src/security/audit-model-refs.ts
@@ -0,0 +1,93 @@
+import { DEFAULT_PROVIDER } from "../agents/defaults.js";
+import { modelKey } from "../agents/model-selection-normalize.js";
+import {
+  buildModelAliasIndex,
+  resolveModelRefFromString,
+} from "../agents/model-selection-shared.js";
+import {
+  resolveAgentModelFallbackValues,
+  resolveAgentModelPrimaryValue,
+} from "../config/model-input.js";
+import type { OpenClawConfig } from "../config/types.openclaw.js";
+
+export type AuditModelRef = { id: string; source: string };
+
+function resolveAuditModelId(
+  cfg: OpenClawConfig,
+  raw: string,
+  aliasIndex: ReturnType<typeof buildModelAliasIndex>,
+): string {
+  const resolved = resolveModelRefFromString({
+    cfg,
+    raw,
+    defaultProvider: DEFAULT_PROVIDER,
+    aliasIndex,
+    allowPluginNormalization: false,
+  })?.ref;
+  return resolved ? modelKey(resolved.provider, resolved.model) : raw;
+}
+
+function addModelRef(params: {
+  out: AuditModelRef[];
+  cfg: OpenClawConfig;
+  aliasIndex: ReturnType<typeof buildModelAliasIndex>;
+  raw: unknown;
+  source: string;
+}): void {
+  if (typeof params.raw !== "string") {
+    return;
+  }
+  const raw = params.raw.trim();
+  if (!raw) {
+    return;
+  }
+  params.out.push({
+    id: resolveAuditModelId(params.cfg, raw, params.aliasIndex),
+    source: params.source,
+  });
+}
+
+export function collectAuditModelRefs(cfg: OpenClawConfig): AuditModelRef[] {
+  const aliasIndex = buildModelAliasIndex({
+    cfg,
+    defaultProvider: DEFAULT_PROVIDER,
+    allowPluginNormalization: false,
+  });
+  const out: AuditModelRef[] = [];
+  const add = (raw: unknown, source: string) => addModelRef({ out, cfg, aliasIndex, raw, source });
+
+  add(resolveAgentModelPrimaryValue(cfg.agents?.defaults?.model), "agents.defaults.model.primary");
+  for (const fallback of resolveAgentModelFallbackValues(cfg.agents?.defaults?.model)) {
+    add(fallback, "agents.defaults.model.fallbacks");
+  }
+  add(
+    resolveAgentModelPrimaryValue(cfg.agents?.defaults?.imageModel),
+    "agents.defaults.imageModel.primary",
+  );
+  for (const fallback of resolveAgentModelFallbackValues(cfg.agents?.defaults?.imageModel)) {
+    add(fallback, "agents.defaults.imageModel.fallbacks");
+  }
+
+  const list = Array.isArray(cfg.agents?.list) ? cfg.agents.list : [];
+  for (const agent of list) {
+    if (!agent || typeof agent !== "object") {
+      continue;
+    }
+    const id =
+      typeof (agent as { id?: unknown }).id === "string" ? (agent as { id: string }).id : "";
+    const model = (agent as { model?: unknown }).model;
+    if (typeof model === "string") {
+      add(model, `agents.list.${id}.model`);
+    } else if (model && typeof model === "object") {
+      add((model as { primary?: unknown }).primary, `agents.list.${id}.model.primary`);
+      const fallbacks = (model as { fallbacks?: unknown }).fallbacks;
+      if (Array.isArray(fallbacks)) {
+        for (const fallback of fallbacks) {
+          add(fallback, `agents.list.${id}.model.fallbacks`);
+        }
+      }
+    }
+  }
+
+  return out;
+}
diff --git a/src/security/audit-small-model-risk.test.ts b/src/security/audit-small-model-risk.test.ts
index 52c7a92541a..7d788b96b9d 100644
--- a/src/security/audit-small-model-risk.test.ts
+++ b/src/security/audit-small-model-risk.test.ts
@@ -45,4 +45,27 @@ describe("security audit small-model risk findings", () => {
       }
     }
   });
+
+  it("resolves configured aliases before parameter-size classification", () => {
+    const [finding] = collectSmallModelRiskFindings({
+      cfg: {
+        agents: {
+          defaults: {
+            model: { primary: "tiny" },
+            models: {
+              "ollama/mistral-8b": { alias: "tiny" },
+            },
+          },
+        },
+        tools: { web: { search: { enabled: true }, fetch: { enabled: true } } },
+        browser: { enabled: true },
+      } satisfies OpenClawConfig,
+      env: {},
+    });
+
+    expect(finding?.checkId).toBe("models.small_params");
+    expect(finding?.detail).toContain("ollama/mistral-8b");
+    expect(finding?.detail).toContain("@ agents.defaults.model.primary");
+    expect(finding?.detail).not.toContain("- tiny");
+  });
 });