@opencode/ai 2.0.20 → 2.0.22

package/dist/cache-policy.js

@@ -36,6 +36,7 @@ const resolve = (policy) => {
 // prefix caching, Gemini's implicit + out-of-band CachedContent). Skip the
 // whole policy pass for these — emitting hints would be harmless but pointless.
 const RESPECTS_INLINE_HINTS = new Set([
+    "alibaba-chat",
     "alibaba-messages",
     "anthropic-messages",
     "anthropic-compatible-messages",
@@ -47,7 +48,21 @@ const RESPECTS_INLINE_HINTS = new Set([
     "zai-coding-messages",
     "bedrock-converse",
     "openrouter",
+    "digitalocean",
 ]);
+// OpenRouter upstreams other than Anthropic and Alibaba Qwen cache without breakpoints. Gemini uses only the last
+// breakpoint, so a conversation-tail breakpoint writes a new cache every step and costs more than none. Qwen ignores
+// breakpoints on tool definitions and caches tools with the system prompt.
+const QWEN = { system: true, messages: { tail: 1 } };
+const openRouterPolicy = (modelID) => {
+    // `~anthropic/claude-sonnet-latest` style IDs are OpenRouter aliases for the latest model in a family.
+    const id = modelID.replace(/^~/, "");
+    if (id.startsWith("anthropic/"))
+        return AUTO;
+    if (id.startsWith("qwen/"))
+        return QWEN;
+    return NONE;
+};
 const makeHint = (ttlSeconds) => ttlSeconds !== undefined ? new CacheHint({ type: "ephemeral", ttlSeconds }) : new CacheHint({ type: "ephemeral" });
 const markLastTool = (tools, hint, budget) => {
     const target = tools.at(-1);
@@ -128,9 +143,13 @@ const countHints = (request) => countToolHints(request.tools) +
 export const applyCachePolicy = (request) => {
     if (!RESPECTS_INLINE_HINTS.has(request.model.route.id))
         return request;
-    if (request.model.route.id === "openrouter" && (request.cache === undefined || request.cache === "auto"))
-        return request;
-    const policy = resolve(request.cache);
+    const policy = request.model.route.id === "openrouter" && (request.cache === undefined || request.cache === "auto")
+        ? openRouterPolicy(request.model.id)
+        : request.model.route.id === "alibaba-chat" && (request.cache === undefined || request.cache === "auto")
+            ? request.model.id.toLowerCase().startsWith("qwen")
+                ? QWEN
+                : NONE
+            : resolve(request.cache);
     if (!policy.tools && !policy.system && !policy.messages)
         return request;
     const hint = makeHint(policy.ttlSeconds);

package/dist/llm.d.ts

@@ -206,6 +206,8 @@ export declare class GenerateObjectResponse<T> {
         readonly reason: {
             readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
             readonly raw?: string | undefined;
+            readonly category?: string | undefined;
+            readonly explanation?: string | undefined;
         };
         readonly index: number;
         readonly providerMetadata?: {
@@ -219,6 +221,8 @@ export declare class GenerateObjectResponse<T> {
         readonly reason: {
             readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
             readonly raw?: string | undefined;
+            readonly category?: string | undefined;
+            readonly explanation?: string | undefined;
         };
         readonly providerMetadata?: {
             readonly [x: string]: {

package/dist/promise.d.ts

@@ -233,6 +233,8 @@ export declare const make: (options?: Options) => {
             readonly reason: {
                 readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
                 readonly raw?: string | undefined;
+                readonly category?: string | undefined;
+                readonly explanation?: string | undefined;
             };
             readonly index: number;
             readonly providerMetadata?: {
@@ -246,6 +248,8 @@ export declare const make: (options?: Options) => {
             readonly reason: {
                 readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
                 readonly raw?: string | undefined;
+                readonly category?: string | undefined;
+                readonly explanation?: string | undefined;
             };
             readonly providerMetadata?: {
                 readonly [x: string]: {
@@ -720,6 +724,8 @@ export declare const ai: {
             readonly reason: {
                 readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
                 readonly raw?: string | undefined;
+                readonly category?: string | undefined;
+                readonly explanation?: string | undefined;
             };
             readonly index: number;
             readonly providerMetadata?: {
@@ -733,6 +739,8 @@ export declare const ai: {
             readonly reason: {
                 readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
                 readonly raw?: string | undefined;
+                readonly category?: string | undefined;
+                readonly explanation?: string | undefined;
             };
             readonly providerMetadata?: {
                 readonly [x: string]: {

package/dist/protocols/alibaba-chat.d.ts

@@ -80,15 +80,18 @@ export declare const protocol: Protocol<{
         })[];
     } | {
         readonly [x: string]: unknown;
-        readonly content: string | null;
+        readonly content: string | readonly {
+            readonly type: "text";
+            readonly text: string;
+            readonly cache_control?: {
+                readonly type: "ephemeral";
+                readonly ttl?: string | undefined;
+            } | undefined;
+        }[] | null;
         readonly role: "assistant";
         readonly reasoning?: string | undefined;
         readonly reasoning_content?: string | undefined;
         readonly reasoning_text?: string | undefined;
-        readonly cache_control?: {
-            readonly type: "ephemeral";
-            readonly ttl?: string | undefined;
-        } | undefined;
         readonly tool_calls?: readonly {
             readonly function: {
                 readonly name: string;
@@ -104,13 +107,16 @@ export declare const protocol: Protocol<{
         }[] | undefined;
         readonly reasoning_details?: unknown;
     } | {
-        readonly content: string;
         readonly role: "tool";
         readonly tool_call_id: string;
-        readonly cache_control?: {
-            readonly type: "ephemeral";
-            readonly ttl?: string | undefined;
-        } | undefined;
+        readonly content: string | readonly {
+            readonly type: "text";
+            readonly text: string;
+            readonly cache_control?: {
+                readonly type: "ephemeral";
+                readonly ttl?: string | undefined;
+            } | undefined;
+        }[];
     })[];
     readonly stream: true;
     readonly max_completion_tokens?: number | undefined;
@@ -180,6 +186,7 @@ export declare const protocol: Protocol<{
     } | null | undefined;
     readonly usage?: {
         readonly [x: string]: unknown;
+        readonly cache_read_input_tokens?: number | null | undefined;
         readonly cached_tokens?: number | null | undefined;
         readonly prompt_tokens?: number | null | undefined;
         readonly completion_tokens?: number | null | undefined;
@@ -190,6 +197,7 @@ export declare const protocol: Protocol<{
             readonly cache_write_tokens?: number | null | undefined;
         } | null | undefined;
         readonly prompt_cache_hit_tokens?: number | null | undefined;
+        readonly cache_created_input_tokens?: number | null | undefined;
         readonly completion_tokens_details?: {
             readonly [x: string]: unknown;
             readonly reasoning_tokens?: number | null | undefined;
@@ -219,6 +227,7 @@ export declare const protocol: Protocol<{
         } | null | undefined;
         readonly usage?: {
             readonly [x: string]: unknown;
+            readonly cache_read_input_tokens?: number | null | undefined;
             readonly cached_tokens?: number | null | undefined;
             readonly prompt_tokens?: number | null | undefined;
             readonly completion_tokens?: number | null | undefined;
@@ -229,6 +238,7 @@ export declare const protocol: Protocol<{
                 readonly cache_write_tokens?: number | null | undefined;
             } | null | undefined;
             readonly prompt_cache_hit_tokens?: number | null | undefined;
+            readonly cache_created_input_tokens?: number | null | undefined;
             readonly completion_tokens_details?: {
                 readonly [x: string]: unknown;
                 readonly reasoning_tokens?: number | null | undefined;

package/dist/protocols/alibaba-chat.js

@@ -2,6 +2,7 @@ import { Effect, Schema } from "effect";
 import { Protocol } from "../route/protocol.js";
 import { OpenAIChat } from "./openai-chat.js";
 import { JsonObject, ProviderShared } from "./shared.js";
+import { cacheControl } from "./utils/cache.js";
 import { OpenResponsesOptions } from "./utils/open-responses-options.js";
 const Options = Schema.Struct({
     reasoningEffort: Schema.optional(OpenResponsesOptions.ReasoningEffort),
@@ -53,7 +54,7 @@ export const protocol = Protocol.make({
         from: Effect.fn("AlibabaChat.fromRequest")(function* (req) {
             const opts = yield* ProviderShared.validateWith(Schema.decodeUnknownEffect(Options))(req.providerOptions ?? {});
             return {
-                ...(yield* OpenAIChat.protocol.body.from(req)),
+                ...(yield* OpenAIChat.fromRequest(req, { cacheControl: cacheControl() })),
                 enable_thinking: opts.enableThinking,
                 // Alibaba also rejects an explicit budget that is not below `max_completion_tokens`.
                 thinking_budget: opts.thinkingBudget === undefined

package/dist/protocols/anthropic-messages.js

@@ -359,6 +359,7 @@ const AnthropicStreamDelta = Schema.Struct({
     signature: Schema.optional(Schema.String),
     stop_reason: optionalNull(Schema.String),
     stop_sequence: optionalNull(Schema.String),
+    stop_details: optionalNull(Schema.Struct({ category: optionalNull(Schema.String), explanation: optionalNull(Schema.String) })),
 });
 const decodeAnthropicStreamDelta = Schema.decodeUnknownOption(AnthropicStreamDelta);
 const AnthropicEvent = Schema.Struct({
@@ -677,16 +678,14 @@ const requireThinkingSignature = (request) => {
 // Mid-conversation system messages became available with Opus 4.8 and version
 // 5 of the other supported Claude families. Treat later family versions as
 // compatible without assuming that every Anthropic Messages model is Claude.
+// Opus 4.8 and every Claude 5 model accept mid-conversation system messages; later versions inherit support.
 const supportsNativeSystemUpdates = (request) => {
-    const match = /(?:^|[./])claude-(fable|haiku|mythos|opus|sonnet)-(\d+)(?:[.-](\d+))?/.exec(String(request.model.id).toLowerCase());
-    if (!match)
+    const version = claudeVersion(String(request.model.id));
+    if (version === undefined)
         return false;
-    const major = Number(match[2]);
-    if (match[1] !== "opus")
-        return major >= 5;
-    if (major !== 4)
-        return major >= 5;
-    return match[3] !== undefined && match[3].length <= 2 && Number(match[3]) >= 8;
+    if (version.family === "opus" && version.major === 4)
+        return version.minor >= 8;
+    return version.major >= 5;
 };
 const endsInServerToolUse = (message) => {
     const last = message.content.at(-1);
@@ -851,6 +850,7 @@ const lowerMessages = Effect.fn("AnthropicMessages.lowerMessages")(function* (re
     }
     return messages;
 });
+// Per-turn effort started with Claude Opus 5 and every Claude 5.1 model; later versions of any family inherit it.
 const supportsEffortUpdates = (model) => {
     const override = model.compatibility?.supportsEffortUpdates;
     if (override !== undefined)
@@ -858,10 +858,8 @@ const supportsEffortUpdates = (model) => {
     const version = claudeVersion(model.id);
     if (version === undefined)
         return false;
-    if (version.family === "opus")
-        return version.major >= 5;
-    if (version.family !== "fable" && version.family !== "mythos")
-        return false;
+    if (version.family === "opus" && version.major >= 5)
+        return true;
     return version.major > 5 || (version.major === 5 && version.minor >= 1);
 };
 const applyThinkingBindingDefault = (model, thinking) => {
@@ -1227,10 +1225,14 @@ const onMessageDelta = (state, event) => {
         const finishMetadata = stopSequence === null || stopSequence === undefined
             ? state.pendingFinish?.providerMetadata
             : providerMetadata(state.providerMetadataKey, { stopSequence });
+        const category = event.delta?.stop_details?.category;
+        const explanation = event.delta?.stop_details?.explanation;
         return {
             reason: {
                 normalized: mapFinishReason(stopReason),
                 raw: stopReason,
+                ...(category ? { category } : {}),
+                ...(explanation ? { explanation } : {}),
             },
             providerMetadata: finishMetadata,
         };

package/dist/protocols/bedrock-converse.js

@@ -236,13 +236,28 @@ const lowerToolResult = Effect.fn("BedrockConverse.lowerToolResult")(function* (
         },
     };
 });
+// Keep Claude and Nova tool-result images inline; put other models' images beside the result.
+const keepToolImagesInline = (id) => id.includes("anthropic.claude-") || id.includes("amazon.nova-");
 const lowerMessages = Effect.fn("BedrockConverse.lowerMessages")(function* (request, breakpoints) {
     const messages = [];
     const documentNames = new Set();
     // Mistral can reject replay IDs even when they satisfy Converse's broader ID syntax.
     const normalizeID = request.model.id.includes("mistral.") ? MistralToolID.normalizer(request) : (id) => id;
     const providerMetadataKey = request.model.route.providerMetadataKey ?? String(request.model.provider);
+    const hoistImages = !keepToolImagesInline(request.model.id);
+    // Bedrock expects parallel tool results before any images hoisted beside them.
+    const pendingImages = [];
+    const flushImages = () => {
+        if (pendingImages.length === 0)
+            return;
+        const previous = messages.at(-1);
+        if (previous?.role === "user")
+            messages[messages.length - 1] = { role: "user", content: [...previous.content, ...pendingImages] };
+        pendingImages.length = 0;
+    };
     for (const message of request.messages) {
+        if (message.role !== "tool")
+            flushImages();
         if (message.role === "system") {
             const part = yield* ProviderShared.wrappedSystemUpdate("Bedrock Converse", message);
             const content = textWithCache(breakpoints, part.text, part.cache);
@@ -317,7 +332,20 @@ const lowerMessages = Effect.fn("BedrockConverse.lowerMessages")(function* (requ
         for (const part of message.content) {
             if (!ProviderShared.supportsContent(part, ["tool-result"]))
                 return yield* ProviderShared.unsupportedContent("Bedrock Converse", "tool", ["tool-result"]);
-            content.push(yield* lowerToolResult(part, documentNames, normalizeID));
+            const result = yield* lowerToolResult(part, documentNames, normalizeID);
+            const images = hoistImages
+                ? result.toolResult.content.filter((item) => "image" in item)
+                : [];
+            const nonImageContent = result.toolResult.content.filter((item) => !("image" in item));
+            content.push(images.length === 0
+                ? result
+                : {
+                    toolResult: {
+                        ...result.toolResult,
+                        content: nonImageContent.length > 0 ? nonImageContent : [{ text: "See attached image." }],
+                    },
+                });
+            pendingImages.push(...images);
             const cachePoint = BedrockCache.block(breakpoints, part.cache);
             if (cachePoint)
                 content.push(cachePoint);
@@ -328,6 +356,7 @@ const lowerMessages = Effect.fn("BedrockConverse.lowerMessages")(function* (requ
         else
             messages.push({ role: "user", content });
     }
+    flushImages();
     return messages;
 });
 // System prompts share the cache-point convention: emit the text block, then

package/dist/protocols/open-responses.d.ts

@@ -1191,6 +1191,8 @@ export declare const step: (state: ParserState, event: NormalizedEvent) => AIErr
     readonly reason: {
         readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
         readonly raw?: string | undefined;
+        readonly category?: string | undefined;
+        readonly explanation?: string | undefined;
     };
     readonly index: number;
     readonly providerMetadata?: {
@@ -1204,6 +1206,8 @@ export declare const step: (state: ParserState, event: NormalizedEvent) => AIErr
     readonly reason: {
         readonly normalized: "length" | "stop" | "tool-calls" | "content-filter" | "error" | "unknown";
         readonly raw?: string | undefined;
+        readonly category?: string | undefined;
+        readonly explanation?: string | undefined;
     };
     readonly providerMetadata?: {
         readonly [x: string]: {