converse-mcp-server 3.0.2 → 3.1.0

package/docs/API.md

@@ -410,8 +410,8 @@ Provide models as plain name strings in the `models` array. Bare names and alias
 | Model | Aliases | Context | Output | Notes |
 |-------|---------|---------|--------|-------|
 | `claude-fable-5` | `fable`, `fable-5` | 1M | 128K | Most capable, adaptive thinking + effort, images, caching, compaction |
-| `claude-opus-4-8` | `opus`, `opus-4.8` | 200K (1M beta) | 128K | Complex reasoning and agentic coding |
-| `claude-opus-4-7`, `claude-opus-4-6` | `opus-4.7`, `opus-4.6` | 200K (1M beta) | 128K | Previous Opus generations |
+| `claude-opus-5` | `opus`, `opus-5` | 1M | 128K | Most capable Opus, adaptive thinking + effort, compaction |
+| `claude-opus-4-8`, `claude-opus-4-7`, `claude-opus-4-6` | `opus-4.8`, `opus-4.7`, `opus-4.6` | 200K (1M beta) | 128K | Previous Opus generations |
 | `claude-opus-4-5-20251101`, `claude-opus-4-1-20250805` | `opus-4.5`, `opus-4.1` | 200K | 64K / 32K | Earlier Opus tiers |
 | `claude-sonnet-4-6` | `sonnet`, `sonnet-4.6` | 200K (1M beta) | 64K | Best speed/intelligence balance, adaptive thinking |
 | `claude-haiku-4-5-20251001` | `haiku`, `haiku-4.5` | 200K | 64K | Fast and intelligent |
@@ -467,7 +467,7 @@ Any other model works via its full `provider/model` slug (e.g. `anthropic/claude
 **Claude** is available through the Claude Agent SDK, using Claude Code CLI authentication instead of an API key:
 
 - **Model**: `claude` (aliases: `claude-sdk`, `claude-code`) — defaults to Claude Fable 5
-- **Model selection**: `claude:fable` (Claude Fable 5) or `claude:opus` (Claude Opus 4.8); unknown `claude:`-prefixed names pass through to the SDK (e.g. `claude:claude-sonnet-4-6`)
+- **Model selection**: `claude:fable` (Claude Fable 5) or `claude:opus` (Claude Opus 5); unknown `claude:`-prefixed names pass through to the SDK (e.g. `claude:claude-sonnet-4-6`)
 - **Authentication**: `claude login` — no `ANTHROPIC_API_KEY` needed
 - **Direct file access**: reads files from the working directory
 - `reasoning_effort` and sampling parameters are managed by the SDK
@@ -501,7 +501,7 @@ agy
 Reach these with the `copilot:` namespace (e.g. `copilot:gpt-5.6-terra`); uses your GitHub Copilot subscription (`gh auth login`) — no API key needed:
 
 - **OpenAI**: `gpt-5.6-sol` (aliases: `gpt-5.6`, `gpt-5`), `gpt-5.6-terra`, `gpt-5.6-luna` (all accept `reasoning_effort`)
-- **Anthropic**: `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-4.8` (aliases: `opus`, `claude`)
+- **Anthropic**: `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-5` (aliases: `opus`, `claude`), `claude-opus-4.8`
 - **Google**: `gemini-3.1-pro-preview` (aliases: `gemini`, `gemini-3.1-pro`), `gemini-3.5-flash` (alias: `gemini-flash`)
 - Any other `copilot:<id>` is forwarded to the Copilot backend verbatim
 
@@ -519,9 +519,9 @@ Use `"auto"` for automatic selection, or specify exact models:
 "z-ai/glm-5.2"             // OpenRouter (full slug)
 "z-ai/glm-5.2:online"      // OpenRouter with web search opt-in
 "fable"                    // Anthropic API (-> claude-fable-5)
-"opus"                     // Anthropic API (-> claude-opus-4-8)
+"opus"                     // Anthropic API (-> claude-opus-5)
 "claude"                   // Claude Agent SDK (-> Claude Fable 5)
-"claude:opus"              // Claude Agent SDK (Claude Opus 4.8)
+"claude:opus"              // Claude Agent SDK (Claude Opus 5)
 "gemini"                   // Antigravity CLI (Gemini 3.1 Pro)
 "copilot:gpt-5.6-terra"    // GitHub Copilot SDK
 ```

package/docs/PROVIDERS.md

@@ -45,8 +45,8 @@ This guide documents all supported AI providers in the Converse MCP Server and t
 - **Environment Variable**: `ANTHROPIC_API_KEY`
 - **Supported Models**:
   - `claude-fable-5` (alias `fable`) - Most capable model for demanding reasoning and long-horizon agentic work (1M context, 128K output)
-  - `claude-opus-4-8` (alias `opus`) - Most capable Opus for complex reasoning and agentic coding (128K output)
-  - `claude-opus-4-7`, `claude-opus-4-6` - Previous Opus generations with adaptive thinking (128K output)
+  - `claude-opus-5` (alias `opus`) - Most capable Opus for complex agentic coding and deep reasoning (1M context, 128K output)
+  - `claude-opus-4-8`, `claude-opus-4-7`, `claude-opus-4-6` - Previous Opus generations with adaptive thinking (128K output)
   - `claude-opus-4-5-20251101`, `claude-opus-4-1-20250805` - Legacy Opus models (64K / 32K output)
   - `claude-sonnet-4-6` (alias `sonnet`) - Best combination of speed and intelligence with adaptive thinking (64K output)
   - `claude-sonnet-4-5-20250929` - Legacy Sonnet (64K output)
@@ -194,7 +194,7 @@ agy
 - **Supported Models**:
   - `claude` (aliases: `claude-sdk`, `claude-code`) - Defaults to Claude Fable 5
   - `claude:fable` - Claude Fable 5 explicitly
-  - `claude:opus` - Claude Opus 4.8
+  - `claude:opus` - Claude Opus 5
   - Other `claude:`-prefixed names pass through to the SDK (e.g. `claude:claude-sonnet-4-6`)
 
 **Key Features:**
@@ -215,7 +215,7 @@ agy
 - **Supported Models** (reach them with the `copilot:` namespace, e.g. `copilot:gpt-5.6-terra`):
   - `copilot` - Uses Copilot's default or env-configured model
   - OpenAI: `gpt-5.6-sol` (aliases: bare `gpt-5.6`, `gpt-5`), `gpt-5.6-terra` (recommended balanced tier), `gpt-5.6-luna`
-  - Anthropic: `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-4.8` (aliases: `opus`, `claude`)
+  - Anthropic: `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-5` (aliases: `opus`, `claude`), `claude-opus-4.8`
   - Google: `gemini-3.1-pro-preview` (aliases: `gemini`, `gemini-3.1-pro`), `gemini-3.5-flash` (alias: `gemini-flash`)
 - **Reasoning**: The `gpt-5.6-sol`/`terra`/`luna` tiers accept `reasoning_effort`.
 - **Explicit pass-through**: Any other `copilot:<id>` model string is forwarded to the Copilot backend verbatim, so IDs outside the curated list still work while the backend accepts them.
@@ -336,10 +336,10 @@ The `models` array always holds plain model-name strings. Each string routes as
 ```text
 "gpt-5.6"                  // OpenAI (keyword match)
 "fable"                    // Anthropic (keyword match -> claude-fable-5)
-"opus"                     // Anthropic (keyword match -> claude-opus-4-8)
+"opus"                     // Anthropic (keyword match -> claude-opus-5)
 "sonnet"                   // Anthropic (keyword match -> claude-sonnet-4-6)
 "claude"                   // Claude Agent SDK (defaults to Claude Fable 5)
-"claude:opus"              // Claude Agent SDK (Claude Opus 4.8)
+"claude:opus"              // Claude Agent SDK (Claude Opus 5)
 "gemini-2.5-pro"           // Google (keyword match)
 "grok-4.5"                 // X.AI (keyword match)
 "mistral-large"            // Mistral (alias -> mistral-large-2512)
@@ -371,7 +371,7 @@ The `models` array always holds plain model-name strings. Each string routes as
 
 ### Model Not Found
 - Use exact model identifiers as listed above
-- Some providers support aliases (e.g., "fable" → "claude-fable-5", "opus" → "claude-opus-4-8")
+- Some providers support aliases (e.g., "fable" → "claude-fable-5", "opus" → "claude-opus-5")
 - Note: bare "claude" routes to the Claude Agent SDK provider, not the Anthropic API
 - Check provider documentation for model availability in your region
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "converse-mcp-server",
-  "version": "3.0.2",
+  "version": "3.1.0",
   "description": "Converse MCP Server - Converse with other LLMs with chat and consensus tools",
   "type": "module",
   "main": "src/index.js",
@@ -93,23 +93,23 @@
     ".env.example"
   ],
   "dependencies": {
-    "@anthropic-ai/claude-agent-sdk": "^0.3.207",
-    "@anthropic-ai/sdk": "^0.111.0",
-    "@github/copilot-sdk": "^1.0.6",
-    "@google/genai": "^2.11.0",
+    "@anthropic-ai/claude-agent-sdk": "^0.3.215",
+    "@anthropic-ai/sdk": "^0.112.3",
+    "@github/copilot-sdk": "^1.0.7",
+    "@google/genai": "^2.12.0",
     "@lydell/node-pty": "1.2.0-beta.12",
-    "@mistralai/mistralai": "^2.4.1",
+    "@mistralai/mistralai": "^2.5.0",
     "@modelcontextprotocol/sdk": "^1.29.0",
-    "@openai/codex-sdk": "^0.144.2",
+    "@openai/codex-sdk": "^0.144.6",
     "ai": "^6.0.208",
     "cors": "^2.8.6",
     "dotenv": "^17.4.2",
     "express": "^5.2.1",
     "lru-cache": "^11.5.2",
     "nanoid": "^5.1.16",
-    "openai": "^6.46.0",
+    "openai": "^6.48.0",
     "p-limit": "^7.3.0",
-    "vite": "^8.1.4"
+    "vite": "^8.1.5"
   },
   "devDependencies": {
     "@vitest/coverage-v8": "^4.1.10",

package/src/providers/anthropic.js

@@ -38,6 +38,32 @@ const SUPPORTED_MODELS = {
       'fable',
     ],
   },
+  'claude-opus-5': {
+    modelName: 'claude-opus-5',
+    friendlyName: 'Claude Opus 5',
+    contextWindow: 1000000, // 1M context by default - no beta header required
+    maxOutputTokens: 128000,
+    supportsStreaming: true,
+    supportsImages: true,
+    supportsWebSearch: false,
+    supportsThinking: true,
+    supportsAdaptiveThinking: true, // Thinking is on by default; adaptive is the only on-mode
+    timeout: 600000,
+    supportsEffort: true,
+    effortGA: true,
+    supportsCompaction: true,
+    description:
+      'Claude Opus 5 - Most capable Opus for complex agentic coding and deep reasoning',
+    aliases: [
+      'claude-opus-5',
+      'claude-5-opus',
+      'opus-5',
+      'opus5',
+      'claude-opus-5.0',
+      'opus',
+      'claude-opus',
+    ],
+  },
   'claude-opus-4-8': {
     modelName: 'claude-opus-4-8',
     friendlyName: 'Claude Opus 4.8',
@@ -56,7 +82,7 @@ const SUPPORTED_MODELS = {
     supports1MContext: true,
     supportsCompaction: true,
     description:
-      'Claude Opus 4.8 - Most capable model for complex reasoning and agentic coding',
+      'Claude Opus 4.8 - Previous Opus generation for complex reasoning and agentic coding',
     aliases: [
       'claude-opus-4-8',
       'claude-4.8-opus',
@@ -66,8 +92,6 @@ const SUPPORTED_MODELS = {
       'opus4.8',
       'opus4-8',
       'claude-opus-4.8',
-      'opus',
-      'claude-opus',
     ],
   },
   'claude-opus-4-7': {
@@ -291,7 +315,7 @@ const THINKING_BUDGETS = {
 };
 
 /**
- * Effort parameter mapping for Opus 4.8, Opus 4.7, Opus 4.6, Sonnet 4.6, and Opus 4.5
+ * Effort parameter mapping for Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, Sonnet 4.6, and Opus 4.5
  * Maps reasoning_effort values to Anthropic's effort parameter values
  */
 const EFFORT_MAP = {

package/src/providers/claude.js

@@ -43,17 +43,17 @@ const SUPPORTED_MODELS = {
     ],
   },
   opus: {
-    modelName: 'claude-opus-4-8',
-    friendlyName: 'Claude Opus 4.8 (via Agent SDK)',
-    contextWindow: 200000,
+    modelName: 'claude-opus-5',
+    friendlyName: 'Claude Opus 5 (via Agent SDK)',
+    contextWindow: 1000000,
     maxOutputTokens: 128000,
     supportsStreaming: true,
     supportsImages: true, // Supported via streaming input mode
     supportsWebSearch: false, // SDK accesses files directly, not web
     timeout: 600000, // 10 minutes
     description:
-      'Claude Opus 4.8 via Agent SDK - requires claude login authentication',
-    aliases: ['claude:opus', 'claude-opus-4-8'],
+      'Claude Opus 5 via Agent SDK - requires claude login authentication',
+    aliases: ['claude:opus', 'claude-opus-5'],
   },
 };
 
@@ -481,7 +481,7 @@ export const claudeProvider = {
         debugLog('[Claude SDK] Using streaming input mode for image support');
       }
 
-      // Resolve requested model (claude/claude:fable -> claude-fable-5, claude:opus -> claude-opus-4-8)
+      // Resolve requested model (claude/claude:fable -> claude-fable-5, claude:opus -> claude-opus-5)
       const sdkModel = resolveSdkModel(model);
       debugLog(`[Claude SDK] Resolved model "${model}" -> "${sdkModel}"`);
 

package/src/providers/codex.js

@@ -27,6 +27,11 @@ const SUPPORTED_MODELS = {
     supportsStreaming: true,
     supportsImages: true, // Codex SDK 0.118+ supports images via --image (local_image input)
     supportsWebSearch: false, // Codex accesses files directly, not web
+    // Reasoning tiers this model's backend actually accepts. GPT-5.6 dropped
+    // 'minimal' and added 'none', while the SDK's ModelReasoningEffort type
+    // still advertises the pre-5.6 set — the backend is the authority, so the
+    // accepted tiers are declared per model and requests are clamped onto them.
+    supportedEfforts: ['none', 'low', 'medium', 'high', 'xhigh'],
     timeout: 600000, // 10 minutes
     description:
       'OpenAI Codex agentic coding assistant with local file access and tool execution (GPT-5.6)',
@@ -212,23 +217,77 @@ async function getThreadIdFromContinuation(
 }
 
 /**
- * Map tool-level reasoning_effort values to Codex SDK's ModelReasoningEffort.
- * Tool enum:  'none' | 'minimal' | 'low' | 'medium' | 'high' | 'max'
- * SDK enum:   'minimal' | 'low' | 'medium' | 'high' | 'xhigh'
+ * Every Codex reasoning tier, weakest to strongest. Used to clamp a requested
+ * tier onto the set a given model actually accepts.
+ */
+const EFFORT_LADDER = ['none', 'minimal', 'low', 'medium', 'high', 'xhigh'];
+
+/**
+ * Tool-level reasoning_effort values translated to their Codex equivalent.
+ * Tool enum: 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'max'
+ */
+const EFFORT_ALIASES = {
+  none: 'none',
+  minimal: 'minimal',
+  low: 'low',
+  medium: 'medium',
+  high: 'high',
+  max: 'xhigh',
+};
+
+/**
+ * Map a tool-level reasoning_effort onto a tier the target model accepts.
+ *
+ * When the requested tier isn't in the model's supported set, the nearest
+ * *stronger* tier wins: nudging 'minimal' up to 'low' keeps reasoning on,
+ * where falling back to 'none' would silently switch it off.
  *
- * The SDK type still lists 'minimal', but the current GPT-5.6-based Codex
- * backend rejects it (5.6 dropped the minimal tier), so 'low' is the floor.
+ * @param {string} effort - Tool-level reasoning_effort value
+ * @param {string[]} [supported] - Tiers the model accepts
+ * @returns {string} A tier from `supported`
+ */
+export function mapReasoningEffort(effort, supported = EFFORT_LADDER) {
+  const desired = EFFORT_ALIASES[effort] || 'medium';
+  if (supported.includes(desired)) {
+    return desired;
+  }
+
+  const rank = EFFORT_LADDER.indexOf(desired);
+  const stronger = EFFORT_LADDER.slice(rank + 1).find((tier) =>
+    supported.includes(tier),
+  );
+  if (stronger) {
+    return stronger;
+  }
+
+  const weaker = EFFORT_LADDER.slice(0, rank)
+    .reverse()
+    .find((tier) => supported.includes(tier));
+  return weaker || 'medium';
+}
+
+/**
+ * Resolve a user-facing model name to its entry in SUPPORTED_MODELS.
+ * @param {string} modelName
+ * @returns {Object|null}
  */
-function mapReasoningEffort(effort) {
-  const mapping = {
-    none: 'low',
-    minimal: 'low',
-    low: 'low',
-    medium: 'medium',
-    high: 'high',
-    max: 'xhigh',
-  };
-  return mapping[effort] || 'medium';
+function findModelConfig(modelName) {
+  const modelNameLower = String(modelName || '').toLowerCase();
+
+  if (SUPPORTED_MODELS[modelNameLower]) {
+    return SUPPORTED_MODELS[modelNameLower];
+  }
+
+  for (const config of Object.values(SUPPORTED_MODELS)) {
+    if (
+      config.aliases &&
+      config.aliases.some((alias) => alias.toLowerCase() === modelNameLower)
+    ) {
+      return config;
+    }
+  }
+
+  return null;
 }
 
 /**
@@ -353,7 +412,16 @@ export const codexProvider = {
       };
 
       if (reasoning_effort) {
-        threadOptions.modelReasoningEffort = mapReasoningEffort(reasoning_effort);
+        const supportedEfforts =
+          findModelConfig(model)?.supportedEfforts ||
+          SUPPORTED_MODELS.codex.supportedEfforts;
+        const mappedEffort = mapReasoningEffort(reasoning_effort, supportedEfforts);
+        threadOptions.modelReasoningEffort = mappedEffort;
+        if (mappedEffort !== EFFORT_ALIASES[reasoning_effort]) {
+          debugLog(
+            `[Codex] reasoning_effort "${reasoning_effort}" not supported by ${model} — using "${mappedEffort}"`,
+          );
+        }
       }
 
       const thread = threadId
@@ -482,23 +550,6 @@ export const codexProvider = {
    * Get model configuration for specific model
    */
   getModelConfig(modelName) {
-    const modelNameLower = modelName.toLowerCase();
-
-    // Check exact match
-    if (SUPPORTED_MODELS[modelNameLower]) {
-      return SUPPORTED_MODELS[modelNameLower];
-    }
-
-    // Check aliases
-    for (const [_name, config] of Object.entries(SUPPORTED_MODELS)) {
-      if (
-        config.aliases &&
-        config.aliases.some((alias) => alias.toLowerCase() === modelNameLower)
-      ) {
-        return config;
-      }
-    }
-
-    return null;
+    return findModelConfig(modelName);
   },
 };

package/src/providers/copilot.js

@@ -104,6 +104,18 @@ const SUPPORTED_MODELS = {
     description: 'Anthropic Claude Sonnet 5 via Copilot subscription',
     aliases: ['sonnet'],
   },
+  'claude-opus-5': {
+    modelName: 'claude-opus-5',
+    friendlyName: 'Claude Opus 5 (via Copilot)',
+    contextWindow: 200000,
+    maxOutputTokens: 32768,
+    supportsStreaming: true,
+    supportsImages: false,
+    supportsWebSearch: false,
+    timeout: 600000,
+    description: 'Anthropic Claude Opus 5 via Copilot subscription',
+    aliases: ['opus', 'claude'],
+  },
   'claude-opus-4.8': {
     modelName: 'claude-opus-4.8',
     friendlyName: 'Claude Opus 4.8 (via Copilot)',
@@ -114,7 +126,7 @@ const SUPPORTED_MODELS = {
     supportsWebSearch: false,
     timeout: 600000,
     description: 'Anthropic Claude Opus 4.8 via Copilot subscription',
-    aliases: ['opus', 'claude'],
+    aliases: [],
   },
 
   // Google models