fix(ollama): bound model-discovery JSON response reads (openclaw#96027)

Alix-007 · QiuYuang · commit 71bf584bca54 · 2026-07-01T18:20:21.000+08:00
* fix(ollama): bound model-discovery JSON response reads The /api/tags and /api/show discovery reads in extensions/ollama/src/provider-models.ts parsed their HTTP responses with an unbounded await response.json(). Ollama base URLs are user-supplied and can point at remote/cloud endpoints, so a hostile or buggy server (or one reachable via SSRF) could stream an unbounded or never-ending JSON body and drive model discovery into OOM. Route both reads through the shared @openclaw/media-core byte-bounded reader (readResponseWithLimit, re-exported via openclaw/plugin-sdk/response-limit-runtime) under a single 16 MiB cap before JSON.parse, cancelling the stream on overflow. Overflow throws a bounded error that the existing fail-soft handlers swallow, so a capped endpoint degrades gracefully: /api/tags returns { reachable: false, models: [] } and /api/show returns {}. Symmetric counterpart to the openclaw#95103/openclaw#95108 response-limit campaign. AI-assisted. * fix(ollama): reuse shared bounded JSON reader for model discovery Replace the local readOllamaDiscoveryJson helper with the shared readProviderJsonResponse (from openclaw/plugin-sdk/provider-http), which already enforces the 16 MiB cap, cancels the stream on overflow, and wraps malformed JSON with the caller label. The /api/tags and /api/show discovery reads now go through it directly while keeping the existing fail-soft handlers ({ reachable: false, models: [] } and {}). Add a focused regression test: when a discovery stream exceeds the JSON byte cap, fetchOllamaModels returns { reachable: false, models: [] }, queryOllamaModelShowInfo returns {}, and the bounded reader cancels the body mid-flight so less than the full advertised stream is read.
diff --git a/extensions/ollama/src/provider-models.test.ts b/extensions/ollama/src/provider-models.test.ts
@@ -5,7 +5,9 @@ import {
   buildOllamaProvider,
   buildOllamaModelDefinition,
   enrichOllamaModelsWithContext,
+  fetchOllamaModels,
   parseOllamaNumCtxParameter,
+  queryOllamaModelShowInfo,
   resetOllamaModelShowInfoCacheForTest,
   resolveOllamaApiBase,
   type OllamaTagModel,
@@ -380,4 +382,57 @@ describe("ollama provider models", () => {
     expect(parseOllamaNumCtxParameter('stop "<|eot_id|>"')).toBeUndefined();
     expect(parseOllamaNumCtxParameter({ num_ctx: 8192 })).toBeUndefined();
   });
+
+  it("fails soft and stops reading when discovery streams exceed the JSON byte cap", async () => {
+    // Larger than the shared 16 MiB readProviderJsonResponse cap so the bounded reader cancels
+    // the stream mid-flight; if the cap were removed the reader would buffer the whole payload.
+    const ONE_MIB = 1024 * 1024;
+    const TOTAL_CHUNKS = 32; // 32 MiB advertised body, double the cap.
+    const chunk = new Uint8Array(ONE_MIB);
+
+    let bytesPulled = 0;
+    let canceled = false;
+    const makeOversizedJsonResponse = (): Response => {
+      bytesPulled = 0;
+      canceled = false;
+      let pulled = 0;
+      const body = new ReadableStream<Uint8Array>({
+        pull(controller) {
+          if (pulled >= TOTAL_CHUNKS) {
+            controller.close();
+            return;
+          }
+          pulled += 1;
+          bytesPulled += chunk.length;
+          controller.enqueue(chunk);
+        },
+        cancel() {
+          canceled = true;
+        },
+      });
+      return new Response(body, {
+        status: 200,
+        headers: { "Content-Type": "application/json" },
+      });
+    };
+
+    vi.stubGlobal(
+      "fetch",
+      vi.fn(async () => makeOversizedJsonResponse()),
+    );
+    const tags = await fetchOllamaModels("http://127.0.0.1:11434");
+    expect(tags).toEqual({ reachable: false, models: [] });
+    expect(canceled).toBe(true);
+    // Only the bounded prefix is pulled, never the full advertised 32 MiB stream.
+    expect(bytesPulled).toBeLessThan(TOTAL_CHUNKS * ONE_MIB);
+
+    vi.stubGlobal(
+      "fetch",
+      vi.fn(async () => makeOversizedJsonResponse()),
+    );
+    const showInfo = await queryOllamaModelShowInfo("http://127.0.0.1:11434", "evil-model:latest");
+    expect(showInfo).toEqual({});
+    expect(canceled).toBe(true);
+    expect(bytesPulled).toBeLessThan(TOTAL_CHUNKS * ONE_MIB);
+  });
 });
diff --git a/extensions/ollama/src/provider-models.ts b/extensions/ollama/src/provider-models.ts
@@ -2,6 +2,7 @@
 import { createHash } from "node:crypto";
 import type { ModelProviderConfig } from "openclaw/plugin-sdk/provider-model-shared";
 import type { ModelDefinitionConfig } from "openclaw/plugin-sdk/provider-onboard";
+import { readProviderJsonResponse } from "openclaw/plugin-sdk/provider-http";
 import { fetchWithSsrFGuard } from "openclaw/plugin-sdk/ssrf-runtime";
 import {
   OLLAMA_DEFAULT_BASE_URL,
@@ -146,11 +147,11 @@ export async function queryOllamaModelShowInfo(
       if (!response.ok) {
         return {};
       }
-      const data = (await response.json()) as {
+      const data = await readProviderJsonResponse<{
         model_info?: Record<string, unknown>;
         capabilities?: unknown;
         parameters?: unknown;
-      };
+      }>(response, "ollama-provider-models.show");
 
       let contextWindow: number | undefined;
       if (data.model_info) {
@@ -314,7 +315,10 @@ export async function fetchOllamaModels(
       if (!response.ok) {
         return { reachable: true, models: [] };
       }
-      const data = (await response.json()) as OllamaTagsResponse;
+      const data = await readProviderJsonResponse<OllamaTagsResponse>(
+        response,
+        "ollama-provider-models.tags",
+      );
       const models = (data.models ?? []).filter((m) => m.name);
       return { reachable: true, models };
     } finally {
