converse-mcp-server 3.6.1 → 3.7.0

package/.env.example

@@ -80,9 +80,9 @@ OPENROUTER_API_KEY=your_openrouter_api_key_here
 # WARNING: Interactive policies may cause hangs in server/headless mode
 # CODEX_APPROVAL_POLICY=never
 
-# Default Codex backend model (default: gpt-6-astra). Per-request override: models: ["codex:<model>"]
-# Options: gpt-6-astra, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5, gpt-5.3-codex-spark
-# CODEX_MODEL=gpt-6-astra
+# Default Codex backend model (default: gpt-6-sol). Per-request override: models: ["codex:<model>"]
+# Options: gpt-6-sol, gpt-6-luna, gpt-6-astra, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5, gpt-5.3-codex-spark
+# CODEX_MODEL=gpt-6-sol
 
 # ============================================
 # Server Configuration

package/README.md

@@ -211,9 +211,12 @@ SUMMARIZATION_MODEL=gpt-5-nano        # Default: gpt-5-nano
 
 ### OpenAI Models
 
-- **gpt-5.6-sol** (default; aliases: `gpt-5.6`, `gpt-5`, `sol`): Flagship GPT-5.6 (1M context, 128K output) - Frontier reasoning, coding, and agentic workflows
-- **gpt-5.6-terra** (alias: `terra`): Lower-cost GPT-5.6 (400K context, 128K output) - Performance competitive with the flagship at half the price
-- **gpt-5.6-luna** (alias: `luna`): Fastest, most affordable GPT-5.6 (400K context, 128K output) - High-volume, latency-sensitive workloads
+- **gpt-6-sol** (default; aliases: `gpt-6`, `gpt-5`, `sol`): Default GPT-6 (1M context, 128K output) - Complex coding and agentic workflows; effort `none`–`max`
+- **gpt-6-luna** (alias: `luna`): Most efficient GPT-6 (1M context, 128K output) - Focused, high-volume tasks; effort `none`–`max`
+- **gpt-6-astra** (alias: `astra`): Frontier GPT-6 flagship (1M context, 128K output) - Hardest end-to-end work; effort `low`–`max` (EXPENSIVE: 5x Sol)
+- **gpt-5.6-sol** (alias: `gpt-5.6`): Previous flagship GPT-5.6 (1M context, 128K output)
+- **gpt-5.6-terra** (alias: `terra`): Lower-cost GPT-5.6 (400K context, 128K output) - Performance competitive with GPT-5.5 at half the price
+- **gpt-5.6-luna**: Fastest, most affordable GPT-5.6 (400K context, 128K output)
 - **gpt-5.4**: Flagship-class reasoning (1M context, 128K output)
 - **gpt-5.4-pro** (alias: `gpt-5-pro`): Maximum-performance reasoning (1M context, 272K output) - Hardest problems, extended compute time (EXPENSIVE)
 - **gpt-5-mini**, **gpt-5-nano**: Faster, cost-efficient GPT-5 tiers (400K context, 128K output)
@@ -247,9 +250,10 @@ SUMMARIZATION_MODEL=gpt-5-nano        # Default: gpt-5-nano
 
 ### Anthropic Models
 
+- **claude-opus-5-5** (default; aliases: `opus`, `opus-5.5`, `claude-opus`): Flagship Opus for complex agentic coding and deep reasoning; thinking always on, effort `low`–`max` (1M context, 128K output)
 - **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 (200K context, 1M via beta, 128K output)
-- **claude-opus-4-7** / **claude-opus-4-6**: Previous Opus generations with adaptive thinking (128K output)
+- **claude-opus-5** (alias: `opus-5`): Previous Opus generation (1M context, 128K output)
+- **claude-opus-4-8** / **claude-opus-4-7** / **claude-opus-4-6**: Earlier Opus generations with adaptive thinking (200K context, 1M via beta, 128K output)
 - **claude-opus-4-5** / **claude-opus-4-1**: Legacy Opus models with extended thinking (64K / 32K output)
 - **claude-sonnet-4-6** (alias: `sonnet`): Best combination of speed and intelligence with adaptive thinking (64K output)
 - **claude-haiku-4-5** (alias: `haiku`): Fast and intelligent for simple queries (64K output)
@@ -281,9 +285,9 @@ Any other model works via its full `provider/model` slug or the `openrouter:` na
 
 ### Codex Models
 
-- **codex**: OpenAI Codex agentic coding assistant (GPT-6 Astra by default)
-  - Pick another backend per request with `codex:<model>` (e.g. `codex:sol`, `codex:gpt-5.6-terra`) or globally with `CODEX_MODEL`; backends: `gpt-6-astra` (alias `astra`), `gpt-5.6-sol` (`sol`), `gpt-5.6-terra` (`terra`), `gpt-5.6-luna` (`luna`), `gpt-5.5`, `gpt-5.3-codex-spark` (`spark`)
-  - `reasoning_effort` maps onto the tiers the chosen backend accepts (GPT-6 Astra: `low` through `max`, no `none`)
+- **codex**: OpenAI Codex agentic coding assistant (GPT-6 Sol by default)
+  - Pick another backend per request with `codex:<model>` (e.g. `codex:luna`, `codex:astra`, `codex:gpt-5.6-terra`) or globally with `CODEX_MODEL`; backends: `gpt-6-sol` (aliases `sol`, `gpt-6`), `gpt-6-luna` (`luna`), `gpt-6-astra` (`astra`), `gpt-5.6-sol`, `gpt-5.6-terra` (`terra`), `gpt-5.6-luna`, `gpt-5.5`, `gpt-5.3-codex-spark` (`spark`)
+  - `reasoning_effort` maps onto the tiers the chosen backend accepts (Sol/Luna: `none` through `max`; GPT-6 Astra: `low` through `max`, no `none`)
   - Thread-based sessions with persistent context
   - Direct filesystem access from working directory
   - Typical response time: 6-20 seconds (longer for complex tasks)
@@ -293,17 +297,17 @@ Any other model works via its full `provider/model` slug or the `openrouter:` na
 ### Claude Agent SDK Models
 
 - **claude** (aliases: `claude-sdk`, `claude-code`): Claude via the Claude Agent SDK
-  - Defaults to Claude Fable 5.1 (`claude-fable-5-1`); `claude:fable` and `claude:fable-5.1` select 5.1, `claude:fable-5` selects 5.0, and `claude:opus` selects Opus 5
+  - Defaults to Claude Opus 5.5 (`claude-opus-5-5`); `claude:opus` selects Opus 5.5, `claude:opus-5` selects Opus 5, `claude:fable` / `claude:fable-5.1` select Fable 5.1, and `claude:fable-5` selects Fable 5.0
   - Uses Claude Code CLI authentication (`claude login`) - no API key needed
   - Direct filesystem access from working directory
   - Unknown `claude:`-prefixed names pass through to the SDK (e.g. `claude:claude-sonnet-4-6`)
 
 ### GitHub Copilot SDK Models
 
-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:
+Reach these with the `copilot:` namespace (e.g. `copilot:gpt-6-sol`); 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 support `reasoning_effort`)
-- **Anthropic**: `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-4.8` (aliases: `opus`, `claude`)
+- **OpenAI**: `gpt-6-sol` (aliases: `gpt-6`, `gpt-5`, `sol`), `gpt-6-luna` (alias: `luna`), `gpt-5.6-sol` (alias: `gpt-5.6`), `gpt-5.6-terra`, `gpt-5.6-luna` (all support `reasoning_effort`)
+- **Anthropic**: `claude-opus-5.5` (aliases: `opus`, `claude`), `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-5`, `claude-opus-4.8`
 - **Google**: `gemini-3.1-pro-preview` (aliases: `gemini`, `gemini-3.1-pro`), `gemini-3.8-flash` (aliases: `gemini-3.8`, `flash-3.8`), `gemini-3.5-flash` (alias: `gemini-flash`)
 - Any other `copilot:<id>` is forwarded to the Copilot backend verbatim
 
@@ -359,7 +363,7 @@ CODEX_API_KEY=your_codex_api_key_here       # Optional if ChatGPT login availabl
 CODEX_SANDBOX_MODE=read-only                 # read-only (default), workspace-write, danger-full-access
 CODEX_SKIP_GIT_CHECK=true                    # true (default), false
 CODEX_APPROVAL_POLICY=never                  # never (default), untrusted, on-failure, on-request
-CODEX_MODEL=gpt-6-astra                      # gpt-6-astra (default), gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5
+CODEX_MODEL=gpt-6-sol                        # gpt-6-sol (default), gpt-6-luna, gpt-6-astra, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5
 ```
 
 ### Configuration Options
@@ -433,12 +437,13 @@ Use `"auto"` for automatic model selection, or specify exact models:
 "deepseek"; // -> deepseek-v4-pro
 "mistral"; // -> mistral-medium-3-5
 "fable"; // -> claude-fable-5 (Anthropic API)
-"opus"; // -> claude-opus-4-8 (Anthropic API)
+"opus"; // -> claude-opus-5-5 (Anthropic API)
 
 // SDK providers (subscription-based, no API key)
-"claude"; // -> Claude Agent SDK (Claude Fable 5.1)
-"claude:opus"; // -> Claude Agent SDK (Claude Opus 4.8)
-"copilot:gpt-5.6-terra"; // -> GitHub Copilot SDK
+"claude"; // -> Claude Agent SDK (Claude Opus 5.5)
+"claude:fable"; // -> Claude Agent SDK (Claude Fable 5.1)
+"codex:luna"; // -> Codex (GPT-6 Luna)
+"copilot:gpt-6-sol"; // -> GitHub Copilot SDK
 ```
 
 **Auto Model Behavior:**
@@ -448,14 +453,14 @@ Use `"auto"` for automatic model selection, or specify exact models:
 
 Provider priority order (subscription-based SDK providers first, then API-key providers):
 
-1. Codex (`codex`)
+1. Codex (`codex` → GPT-6 Sol)
 2. Gemini via Antigravity CLI (`gemini` → Gemini 3.8 Flash, `gemini:pro`)
-3. Claude Agent SDK (`claude` → Claude Fable 5.1)
+3. Claude Agent SDK (`claude` → Claude Opus 5.5)
 4. Copilot (`copilot`)
-5. OpenAI (`gpt-5.6`)
+5. OpenAI (`gpt-6` → GPT-6 Sol)
 6. Google (`gemini-pro`)
 7. XAI (`grok-4.5`)
-8. Anthropic (`claude-sonnet-4-20250514`)
+8. Anthropic (`claude-opus-5-5`)
 9. Mistral (`mistral-medium-3-5`)
 10. DeepSeek (`deepseek-v4-pro`)
 11. OpenRouter (`z-ai/glm-5.2`)

package/docs/API.md

@@ -374,9 +374,12 @@ Provide models as plain name strings in the `models` array. Bare names and alias
 
 | Model | Aliases | Context | Output | Notes |
 |-------|---------|---------|--------|-------|
-| `gpt-5.6-sol` | `gpt-5.6`, `gpt-5`, `sol` | 1M | 128K | Flagship, default OpenAI model |
-| `gpt-5.6-terra` | `terra` | 400K | 128K | Lower-cost flagship-class tier |
-| `gpt-5.6-luna` | `luna` | 400K | 128K | Fastest, most affordable tier |
+| `gpt-6-sol` | `gpt-6`, `gpt-5`, `sol` | 1M | 128K | Default OpenAI model; effort `none`–`max` |
+| `gpt-6-luna` | `luna` | 1M | 128K | Most efficient GPT-6; effort `none`–`max` |
+| `gpt-6-astra` | `astra` | 1M | 128K | Frontier flagship (expensive); effort `low`–`max`, no `none` |
+| `gpt-5.6-sol` | `gpt-5.6` | 1M | 128K | Previous flagship |
+| `gpt-5.6-terra` | `terra` | 400K | 128K | Lower-cost GPT-5.6 tier |
+| `gpt-5.6-luna` | — | 400K | 128K | Fastest GPT-5.6 tier |
 | `gpt-5.4` | — | 1M | 128K | Flagship-class reasoning |
 | `gpt-5.4-pro` | `gpt-5-pro` | 1M | 272K | Maximum performance (expensive) |
 | `gpt-5-mini`, `gpt-5-nano` | — | 400K | 128K | Fast, cost-efficient tiers |
@@ -410,14 +413,15 @@ Provide models as plain name strings in the `models` array. Bare names and alias
 
 | Model | Aliases | Context | Output | Notes |
 |-------|---------|---------|--------|-------|
+| `claude-opus-5-5` | `opus`, `opus-5.5`, `claude-opus` | 1M | 128K | Default. Flagship Opus, always-on adaptive thinking + effort (no compaction yet) |
 | `claude-fable-5` | `fable`, `fable-5` | 1M | 128K | Most capable, adaptive thinking + effort, images, caching, compaction |
-| `claude-opus-5` | `opus`, `opus-5` | 1M | 128K | Most capable Opus, adaptive thinking + effort, compaction |
+| `claude-opus-5` | `opus-5` | 1M | 128K | Previous 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 |
 
-Models with adaptive thinking control depth via `reasoning_effort`, which is passed by name to Anthropic's `effort` parameter and clamped to what each model accepts: Fable 5, Opus 5, Opus 4.8, and Opus 4.7 take `low`–`max`; Opus 4.6 and Sonnet 4.6 lack `xhigh` (it becomes `max`); Opus 4.5 tops out at `high`. `none` and `minimal` become `low` everywhere. System prompts are automatically cached for 1 hour; cache stats appear in response metadata as `cache_creation_input_tokens` / `cache_read_input_tokens`.
+Models with adaptive thinking control depth via `reasoning_effort`, which is passed by name to Anthropic's `effort` parameter and clamped to what each model accepts: Opus 5.5, Fable 5, Opus 5, Opus 4.8, and Opus 4.7 take `low`–`max`; Opus 4.6 and Sonnet 4.6 lack `xhigh` (it becomes `max`); Opus 4.5 tops out at `high`. `none` and `minimal` become `low` everywhere. System prompts are automatically cached for 1 hour; cache stats appear in response metadata as `cache_creation_input_tokens` / `cache_read_input_tokens`.
 
 ### Mistral Models
 
@@ -456,20 +460,20 @@ Any other model works via its full `provider/model` slug (e.g. `anthropic/claude
 
 **Codex** is an agentic coding assistant with direct filesystem access:
 
-- **Model**: `codex` (underlying model: GPT-6 Astra by default)
-- **Backend selection**: `codex:<model>` per request (e.g. `codex:astra`, `codex:sol`, `codex:gpt-5.6-terra`), or `CODEX_MODEL` globally; unknown names pass through to the CLI verbatim
+- **Model**: `codex` (underlying model: GPT-6 Sol by default)
+- **Backend selection**: `codex:<model>` per request (e.g. `codex:luna`, `codex:astra`, `codex:gpt-5.6-terra`), or `CODEX_MODEL` globally; `sol`/`luna`/`gpt-6` name the GPT-6 tiers, the GPT-5.6 tiers are reached by full slug; unknown names pass through to the CLI verbatim
 - **Thread-based sessions**: persistent conversation history via `continuation_id` in `chat` mode
 - **Direct file access**: reads files from the working directory (paths relative to `CLIENT_CWD`)
 - **Response times**: 6-20 seconds typical (complex tasks may take minutes)
 - **Authentication**: ChatGPT login OR `CODEX_API_KEY` (NOT `OPENAI_API_KEY`)
-- `reasoning_effort` is clamped onto the tiers the chosen backend accepts (GPT-6 Astra: `low`–`max`, no `none`; GPT-5.6: `none`–`max`); web search is not applicable — Codex manages its own execution
+- `reasoning_effort` is clamped onto the tiers the chosen backend accepts (GPT-6 Sol/Luna and GPT-5.6: `none`–`max`; GPT-6 Astra: `low`–`max`, no `none`); web search is not applicable — Codex manages its own execution
 
 ### Claude Agent SDK (subscription)
 
 **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 5.1 (`claude-fable-5-1`)
-- **Model selection**: `claude:fable` or `claude:fable-5.1` (Claude Fable 5.1), `claude:fable-5` (Claude Fable 5.0), or `claude:opus` (Claude Opus 5); unknown `claude:`-prefixed names pass through to the SDK (e.g. `claude:claude-sonnet-4-6`)
+- **Model**: `claude` (aliases: `claude-sdk`, `claude-code`) — defaults to Claude Opus 5.5 (`claude-opus-5-5`)
+- **Model selection**: `claude:opus` or `claude:opus-5.5` (Claude Opus 5.5), `claude:opus-5` (Claude Opus 5), `claude:fable` or `claude:fable-5.1` (Claude Fable 5.1), `claude:fable-5` (Claude Fable 5.0); 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**: `reasoning_effort` maps to the SDK's `effort` option: `low`, `medium`, `high`, `xhigh`, or `max`; `none` and `minimal` become `low`. Omitting it retains the SDK default.
@@ -502,10 +506,10 @@ agy
 
 ### GitHub Copilot SDK (subscription)
 
-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:
+Reach these with the `copilot:` namespace (e.g. `copilot:gpt-6-sol`); 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-5` (aliases: `opus`, `claude`), `claude-opus-4.8`
+- **OpenAI**: `gpt-6-sol` (aliases: `gpt-6`, `gpt-5`, `sol`), `gpt-6-luna` (alias: `luna`), `gpt-5.6-sol` (alias: `gpt-5.6`), `gpt-5.6-terra`, `gpt-5.6-luna` (all accept `reasoning_effort`)
+- **Anthropic**: `claude-opus-5.5` (aliases: `opus`, `claude`), `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-5`, `claude-opus-4.8`
 - **Google**: `gemini-3.1-pro-preview` (aliases: `gemini`, `gemini-3.1-pro`), `gemini-3.8-flash` (aliases: `gemini-3.8`, `flash-3.8`), `gemini-3.5-flash` (alias: `gemini-flash`)
 - Any other `copilot:<id>` is forwarded to the Copilot backend verbatim
 
@@ -515,7 +519,7 @@ Use `"auto"` for automatic selection, or specify exact models:
 
 ```text
 "auto"                     // First available provider (chat); first 3 (consensus)
-"gpt-5.6"                  // OpenAI flagship
+"gpt-6"                    // OpenAI flagship (-> gpt-6-sol)
 "gemini-2.5-flash"         // Google API
 "grok-4.5"                 // X.AI
 "deepseek"                 // DeepSeek (-> deepseek-v4-pro)
@@ -523,11 +527,12 @@ 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-5)
-"claude"                   // Claude Agent SDK (-> Claude Fable 5.1)
-"claude:opus"              // Claude Agent SDK (Claude Opus 5)
+"opus"                     // Anthropic API (-> claude-opus-5-5)
+"claude"                   // Claude Agent SDK (-> Claude Opus 5.5)
+"claude:fable"             // Claude Agent SDK (Claude Fable 5.1)
+"codex:luna"               // Codex (GPT-6 Luna)
 "gemini"                   // Antigravity CLI (Gemini 3.8 Flash)
-"copilot:gpt-5.6-terra"    // GitHub Copilot SDK
+"copilot:gpt-6-sol"        // GitHub Copilot SDK
 ```
 
 **Auto behavior:**
@@ -554,7 +559,7 @@ Control Codex behavior through environment variables:
 - **`CODEX_SANDBOX_MODE`** — filesystem access: `read-only` (default), `workspace-write`, `danger-full-access` (containers only)
 - **`CODEX_SKIP_GIT_CHECK`** — `true` (default) works in any directory; `false` requires a Git repository
 - **`CODEX_APPROVAL_POLICY`** — `never` (default, recommended for servers), `untrusted`, `on-failure`, `on-request`
-- **`CODEX_MODEL`** — underlying model for Codex sessions (default: `gpt-5.6-sol`)
+- **`CODEX_MODEL`** — underlying model for Codex sessions (default: `gpt-6-sol`)
 - **`CODEX_API_KEY`** — optional API key for headless deployments (alternative to ChatGPT login)
 
 **Example (.env):**
@@ -563,7 +568,7 @@ CODEX_API_KEY=your_codex_api_key_here
 CODEX_SANDBOX_MODE=read-only
 CODEX_SKIP_GIT_CHECK=true
 CODEX_APPROVAL_POLICY=never
-CODEX_MODEL=gpt-5.6-sol
+CODEX_MODEL=gpt-6-sol
 ```
 
 ## Context Processing

package/docs/PROVIDERS.md

@@ -9,9 +9,12 @@ This guide documents all supported AI providers in the Converse MCP Server and t
 - **Get Key**: [platform.openai.com/api-keys](https://platform.openai.com/api-keys)
 - **Environment Variable**: `OPENAI_API_KEY`
 - **Supported Models**:
-  - `gpt-5.6-sol` (aliases: `gpt-5.6`, `gpt-5`, `sol`) - Flagship GPT-5.6, the default OpenAI model
+  - `gpt-6-sol` (aliases: `gpt-6`, `gpt-5`, `sol`) - Default GPT-6 and the default OpenAI model (1M context, 128K output; effort `none`–`max`)
+  - `gpt-6-luna` (alias: `luna`) - Most efficient GPT-6 for focused, high-volume tasks (1M context, 128K output; effort `none`–`max`)
+  - `gpt-6-astra` (alias: `astra`) - Frontier GPT-6 flagship, 5x the Sol price (1M context, 128K output; effort `low`–`max`)
+  - `gpt-5.6-sol` (alias: `gpt-5.6`) - Previous flagship GPT-5.6
   - `gpt-5.6-terra` (alias: `terra`) - Lower-cost GPT-5.6, competitive with GPT-5.5
-  - `gpt-5.6-luna` (alias: `luna`) - Fastest, most affordable GPT-5.6
+  - `gpt-5.6-luna` - Fastest, most affordable GPT-5.6
   - `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.4-nano`, `gpt-5.4-pro`, `gpt-5-mini`, `gpt-5-nano` - GPT-5.4/GPT-5 family
   - `o3`, `o3-pro`, `o4-mini` - Advanced reasoning models
   - `gpt-4.1` - Large context (1M tokens)
@@ -45,9 +48,10 @@ This guide documents all supported AI providers in the Converse MCP Server and t
 - **Get Key**: [console.anthropic.com](https://console.anthropic.com/)
 - **Environment Variable**: `ANTHROPIC_API_KEY`
 - **Supported Models**:
+  - `claude-opus-5-5` (aliases `opus`, `opus-5.5`, `claude-opus`) - Default. Flagship Opus for complex agentic coding and deep reasoning; thinking is always on, effort `low`–`max` (1M context, 128K output)
   - `claude-fable-5` (alias `fable`) - Most capable model for demanding reasoning and long-horizon agentic work (1M context, 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-5` (alias `opus-5`) - Previous Opus generation (1M context, 128K output)
+  - `claude-opus-4-8`, `claude-opus-4-7`, `claude-opus-4-6` - Earlier 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)
@@ -101,11 +105,11 @@ This guide documents all supported AI providers in the Converse MCP Server and t
   - `CODEX_SANDBOX_MODE` - Filesystem access control (default: read-only)
   - `CODEX_SKIP_GIT_CHECK` - Skip Git repository validation (default: true)
   - `CODEX_APPROVAL_POLICY` - Command approval behavior (default: never)
-  - `CODEX_MODEL` - Underlying model for Codex sessions (default: gpt-6-astra; e.g. gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5)
+  - `CODEX_MODEL` - Underlying model for Codex sessions (default: gpt-6-sol; e.g. gpt-6-luna, gpt-6-astra, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5)
 - **Supported Models**:
-  - `codex` - OpenAI Codex agentic coding assistant (GPT-6 Astra by default)
-  - `codex:<model>` - Same, with an explicit backend: `codex:astra`, `codex:sol`, `codex:terra`, `codex:luna`, `codex:gpt-5.5`, `codex:spark`, or any slug the Codex CLI knows
-  - `reasoning_effort` is clamped onto what the backend accepts (GPT-6 Astra: `low`–`max`, no `none`)
+  - `codex` - OpenAI Codex agentic coding assistant (GPT-6 Sol by default)
+  - `codex:<model>` - Same, with an explicit backend: `codex:sol`, `codex:luna`, `codex:astra` (GPT-6), `codex:gpt-5.6-sol`, `codex:terra`, `codex:gpt-5.6-luna`, `codex:gpt-5.5`, `codex:spark`, or any slug the Codex CLI knows
+  - `reasoning_effort` is clamped onto what the backend accepts (Sol/Luna: `none`–`max`; GPT-6 Astra: `low`–`max`, no `none`)
   - Thread-based sessions with persistent context
   - Direct filesystem access from working directory
   - Typical response time: 6-20 seconds (longer for complex tasks)
@@ -195,10 +199,11 @@ agy
 - **Setup Required**: Authenticate once with `claude login` (Claude Code CLI)
 - **Environment Variables**: None (uses Claude Code credentials)
 - **Supported Models**:
-  - `claude` (aliases: `claude-sdk`, `claude-code`) - Defaults to Claude Fable 5.1 (`claude-fable-5-1`)
-  - `claude:fable` or `claude:fable-5.1` - Claude Fable 5.1 explicitly
+  - `claude` (aliases: `claude-sdk`, `claude-code`) - Defaults to Claude Opus 5.5 (`claude-opus-5-5`)
+  - `claude:opus` or `claude:opus-5.5` - Claude Opus 5.5 explicitly
+  - `claude:opus-5` - Claude Opus 5 (`claude-opus-5`)
+  - `claude:fable` or `claude:fable-5.1` - Claude Fable 5.1 (`claude-fable-5-1`)
   - `claude:fable-5` - Claude Fable 5.0 (`claude-fable-5`)
-  - `claude:opus` - Claude Opus 5
   - Other `claude:`-prefixed names pass through to the SDK (e.g. `claude:claude-sonnet-4-6`)
 
 **Key Features:**
@@ -218,12 +223,12 @@ agy
 - **Authentication**: GitHub Copilot subscription via the Copilot CLI (`gh auth login` with an active Copilot subscription) — no API key needed
 - **Setup Required**: Authenticate the GitHub CLI and ensure your account has an active Copilot subscription
 - **Environment Variables**: None
-- **Supported Models** (reach them with the `copilot:` namespace, e.g. `copilot:gpt-5.6-terra`):
+- **Supported Models** (reach them with the `copilot:` namespace, e.g. `copilot:gpt-6-sol`):
   - `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-5` (aliases: `opus`, `claude`), `claude-opus-4.8`
+  - OpenAI: `gpt-6-sol` (aliases: bare `gpt-6`, `gpt-5`, `sol`), `gpt-6-luna` (alias: `luna`), `gpt-5.6-sol` (alias: `gpt-5.6`), `gpt-5.6-terra`, `gpt-5.6-luna`
+  - Anthropic: `claude-opus-5.5` (aliases: `opus`, `claude`; Copilot Pro+/Max/Business/Enterprise), `claude-fable-5` (alias: `fable`), `claude-sonnet-5` (alias: `sonnet`), `claude-opus-5`, `claude-opus-4.8`
   - Google: `gemini-3.1-pro-preview` (aliases: `gemini`, `gemini-3.1-pro`), `gemini-3.8-flash` (aliases: `gemini-3.8`, `flash-3.8`), `gemini-3.5-flash` (alias: `gemini-flash`)
-- **Reasoning**: The `gpt-5.6-sol`/`terra`/`luna` tiers accept `reasoning_effort`.
+- **Reasoning**: The GPT-6 and GPT-5.6 tiers accept `reasoning_effort` (clamped onto Copilot's `low`–`xhigh`).
 - **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.
 
 **Key Features:**
@@ -340,12 +345,12 @@ When using the chat tool in any mode, specify models using their identifiers:
 The `models` array always holds plain model-name strings. Each string routes as follows:
 
 ```text
-"gpt-5.6"                  // OpenAI (keyword match)
+"gpt-6"                    // OpenAI (keyword match -> gpt-6-sol)
 "fable"                    // Anthropic (keyword match -> claude-fable-5)
-"opus"                     // Anthropic (keyword match -> claude-opus-5)
+"opus"                     // Anthropic (keyword match -> claude-opus-5-5)
 "sonnet"                   // Anthropic (keyword match -> claude-sonnet-4-6)
-"claude"                   // Claude Agent SDK (defaults to Claude Fable 5.1)
-"claude:opus"              // Claude Agent SDK (Claude Opus 5)
+"claude"                   // Claude Agent SDK (defaults to Claude Opus 5.5)
+"claude:fable"             // Claude Agent SDK (Claude Fable 5.1)
 "gemini-2.5-pro"           // Google (keyword match)
 "grok-4.5"                 // X.AI (keyword match)
 "mistral-large"            // Mistral (alias -> mistral-large-2512)
@@ -377,7 +382,7 @@ 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-5")
+- Some providers support aliases (e.g., "fable" → "claude-fable-5", "opus" → "claude-opus-5-5")
 - 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": "3.6.1",
+  "version": "3.7.0",
   "description": "Converse MCP Server - Converse with other LLMs with chat and consensus tools",
   "type": "module",
   "main": "src/index.js",
@@ -93,28 +93,28 @@
     ".env.example"
   ],
   "dependencies": {
-    "@anthropic-ai/claude-agent-sdk": "^0.3.263",
-    "@anthropic-ai/sdk": "^0.124.0",
-    "@github/copilot-sdk": "^1.0.13",
-    "@google/genai": "^2.21.0",
+    "@anthropic-ai/claude-agent-sdk": "^0.3.276",
+    "@anthropic-ai/sdk": "^0.126.0",
+    "@github/copilot-sdk": "^1.0.14",
+    "@google/genai": "^2.23.0",
     "@lydell/node-pty": "1.2.0-beta.15",
-    "@mistralai/mistralai": "^2.6.4",
+    "@mistralai/mistralai": "^2.7.0",
     "@modelcontextprotocol/sdk": "^1.30.0",
-    "@openai/codex-sdk": "^0.153.4",
+    "@openai/codex-sdk": "^0.155.0",
     "cors": "^2.8.6",
     "dotenv": "^17.4.2",
     "express": "^5.2.1",
     "lru-cache": "^11.5.2",
     "nanoid": "^6.0.1",
-    "openai": "^7.10.0",
+    "openai": "^7.18.0",
     "p-limit": "^7.3.2",
-    "vite": "^8.2.2"
+    "vite": "^8.3.0"
   },
   "devDependencies": {
-    "@vitest/coverage-v8": "^5.0.0",
+    "@vitest/coverage-v8": "^5.0.1",
     "cross-env": "^10.1.0",
     "eslint": "^10.10.0",
     "rimraf": "^6.1.3",
-    "vitest": "^5.0.0"
+    "vitest": "^5.0.1"
   }
 }

package/src/config.js

@@ -280,9 +280,9 @@ const CONFIG_SCHEMA = {
     },
     CODEX_MODEL: {
       type: 'string',
-      default: 'gpt-6-astra',
+      default: 'gpt-6-sol',
       description:
-        'Default Codex backend model (e.g., gpt-6-astra, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5)',
+        'Default Codex backend model (e.g., gpt-6-sol, gpt-6-luna, gpt-6-astra, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5)',
     },
 
     // Copilot configuration
@@ -295,7 +295,7 @@ const CONFIG_SCHEMA = {
       type: 'string',
       required: false,
       description:
-        'Default model for Copilot SDK sessions (e.g., gpt-5.6-sol, claude-sonnet-5)',
+        'Default model for Copilot SDK sessions (e.g., gpt-6-sol, claude-opus-5.5, claude-sonnet-5)',
     },
     COPILOT_CLI_PATH: {
       type: 'string',

package/src/providers/anthropic.js

@@ -21,6 +21,37 @@ const EFFORT_TIERS_LEGACY = ['low', 'medium', 'high'];
 
 // Define supported Claude models with their capabilities
 const SUPPORTED_MODELS = {
+  'claude-opus-5-5': {
+    modelName: 'claude-opus-5-5',
+    friendlyName: 'Claude Opus 5.5',
+    contextWindow: 1000000, // 1M context by default - no beta header required
+    maxOutputTokens: 128000,
+    supportsStreaming: true,
+    supportsImages: true,
+    supportsWebSearch: false,
+    supportsThinking: true,
+    supportsAdaptiveThinking: true, // Thinking is always on and cannot be disabled
+    timeout: 1800000,
+    supportsEffort: true,
+    effortGA: true,
+    effortTiers: EFFORT_TIERS_FULL,
+    // Absent from Anthropic's compaction compatibility list, unlike Opus 5
+    supportsCompaction: false,
+    description:
+      'Claude Opus 5.5 - Flagship Opus for complex agentic coding and deep reasoning; matches Fable 5.1 on most work at Opus pricing',
+    aliases: [
+      'claude-opus-5-5',
+      'claude-opus-5.5',
+      'claude-5.5-opus',
+      'claude-5-5-opus',
+      'opus-5.5',
+      'opus-5-5',
+      'opus5.5',
+      'opus5-5',
+      'opus',
+      'claude-opus',
+    ],
+  },
   'claude-fable-5': {
     modelName: 'claude-fable-5',
     friendlyName: 'Claude Fable 5',
@@ -63,15 +94,13 @@ const SUPPORTED_MODELS = {
     effortTiers: EFFORT_TIERS_FULL,
     supportsCompaction: true,
     description:
-      'Claude Opus 5 - Most capable Opus for complex agentic coding and deep reasoning',
+      'Claude Opus 5 - Previous Opus generation 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': {
@@ -567,7 +596,7 @@ export const anthropicProvider = {
    */
   async invoke(messages, options = {}) {
     const {
-      model = 'claude-3-5-sonnet-20241022',
+      model = 'claude-opus-5-5',
       maxTokens = null,
       stream = false,
       reasoning_effort = 'medium',

package/src/providers/claude.js

@@ -17,15 +17,15 @@ import { debugLog, debugError } from '../utils/console.js';
 import { ProviderError, ErrorCodes, StopReasons } from './interface.js';
 import { clampReasoningEffort } from '../utils/reasoningEffort.js';
 
-// Default underlying model when the request is just "claude" (or "claude:fable")
-const DEFAULT_SDK_MODEL = 'claude-fable-5-1';
+// Default underlying model when the request is just "claude" (or "claude:opus")
+const DEFAULT_SDK_MODEL = 'claude-opus-5-5';
 const SDK_EFFORT_TIERS = ['low', 'medium', 'high', 'xhigh', 'max'];
 
 // Supported Claude SDK models with their configurations
 const SUPPORTED_MODELS = {
-  'fable-5': {
-    modelName: 'claude-fable-5',
-    friendlyName: 'Claude Fable 5 (via Agent SDK)',
+  opus: {
+    modelName: 'claude-opus-5-5',
+    friendlyName: 'Claude Opus 5.5 (via Agent SDK)',
     contextWindow: 1000000,
     maxOutputTokens: 128000,
     supportsStreaming: true,
@@ -33,8 +33,31 @@ const SUPPORTED_MODELS = {
     supportsWebSearch: false, // SDK accesses files directly, not web
     timeout: 1800000, // 30 minutes
     description:
-      'Claude Fable 5 via Agent SDK - requires claude login authentication',
-    aliases: ['claude-fable-5'],
+      'Claude Opus 5.5 via Agent SDK (default) - requires claude login authentication',
+    aliases: [
+      'claude',
+      'claude-sdk',
+      'claude-code',
+      'claude:opus',
+      'claude-opus',
+      'claude-opus-5-5',
+      'claude-opus-5.5',
+      'opus-5-5',
+      'opus-5.5',
+    ],
+  },
+  'opus-5': {
+    modelName: 'claude-opus-5',
+    friendlyName: 'Claude Opus 5 (via Agent SDK)',
+    contextWindow: 1000000,
+    maxOutputTokens: 128000,
+    supportsStreaming: true,
+    supportsImages: true,
+    supportsWebSearch: false,
+    timeout: 1800000,
+    description:
+      'Claude Opus 5 via Agent SDK - requires claude login authentication',
+    aliases: ['claude-opus-5'],
   },
   fable: {
     modelName: 'claude-fable-5-1',
@@ -46,11 +69,8 @@ const SUPPORTED_MODELS = {
     supportsWebSearch: false,
     timeout: 1800000,
     description:
-      'Claude Fable 5.1 via Agent SDK (default) - requires claude login authentication',
+      'Claude Fable 5.1 via Agent SDK - requires claude login authentication',
     aliases: [
-      'claude',
-      'claude-sdk',
-      'claude-code',
       'claude:fable',
       'claude-fable',
       'claude-fable-5-1',
@@ -59,18 +79,18 @@ const SUPPORTED_MODELS = {
       'fable-5.1',
     ],
   },
-  opus: {
-    modelName: 'claude-opus-5',
-    friendlyName: 'Claude Opus 5 (via Agent SDK)',
+  'fable-5': {
+    modelName: 'claude-fable-5',
+    friendlyName: 'Claude Fable 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: 1800000, // 30 minutes
+    supportsImages: true,
+    supportsWebSearch: false,
+    timeout: 1800000,
     description:
-      'Claude Opus 5 via Agent SDK - requires claude login authentication',
-    aliases: ['claude:opus', 'claude-opus-5'],
+      'Claude Fable 5 via Agent SDK - requires claude login authentication',
+    aliases: ['claude-fable-5'],
   },
 };
 
@@ -133,7 +153,7 @@ function findModelConfig(modelName) {
   if (name.toLowerCase().startsWith('claude:')) {
     name = name.slice('claude:'.length).trim();
   }
-  if (!name) return SUPPORTED_MODELS.fable;
+  if (!name) return SUPPORTED_MODELS.opus;
 
   const nameLower = name.toLowerCase();
 
@@ -155,8 +175,8 @@ function findModelConfig(modelName) {
 
 /**
  * Resolve the requested model to the underlying SDK model ID.
- * - "claude" (and bare "claude:") defaults to Claude Fable 5.1
- * - "claude:fable" / "claude:opus" select the specific model
+ * - "claude" (and bare "claude:") defaults to Claude Opus 5.5
+ * - "claude:opus" / "claude:opus-5" / "claude:fable" / "claude:fable-5" select the specific model
  * - Unknown names are passed through (after prefix stripping) so users can
  *   target any model ID the Agent SDK accepts (e.g. "claude:claude-sonnet-4-6")
  */
@@ -624,7 +644,7 @@ export const claudeProvider = {
 
   /**
    * Get model configuration for specific model
-   * Handles claude: prefixed names (e.g. "claude:opus", "claude:fable")
+   * Handles claude: prefixed names (e.g. "claude:opus", "claude:fable", "claude:opus-5")
    */
   getModelConfig(modelName) {
     return findModelConfig(modelName);

package/src/providers/codex.js

@@ -25,9 +25,13 @@ import {
  * Backend models Codex can run, keyed by the slug passed to the CLI as
  * --model. The reasoning tiers are the ones each model's API accepts, verified
  * against the API's own rejection messages (gpt-6-astra: "Supported values
- * are: 'low', 'medium', 'high', 'xhigh', and 'max'"). The SDK's
- * ModelReasoningEffort type is the union across models, so the backend is the
- * authority and requests are clamped per model.
+ * are: 'low', 'medium', 'high', 'xhigh', and 'max'"; the Sol/Luna tiers of
+ * both generations accept 'none' as well). The SDK's ModelReasoningEffort
+ * type is the union across models, so the backend is the authority and
+ * requests are clamped per model.
+ *
+ * Bare tier names (sol, luna) and the bare generation (gpt-6) point at the
+ * current generation; the GPT-5.6 tiers stay reachable by full slug.
  *
  * Codex also exposes 'ultra' above 'max', but that tier turns on automatic
  * sub-agent delegation — a change in how the run executes, not just how deep
@@ -35,23 +39,33 @@ import {
  * and nothing at the tool level can select it.
  */
 const CODEX_BACKEND_MODELS = {
+  'gpt-6-sol': {
+    aliases: ['sol', 'gpt-6', 'gpt6', 'gpt6-sol', 'gpt-6-codex'],
+    contextWindow: 272000,
+    supportedEfforts: ['none', 'low', 'medium', 'high', 'xhigh', 'max'],
+  },
+  'gpt-6-luna': {
+    aliases: ['luna', 'gpt6-luna'],
+    contextWindow: 272000,
+    supportedEfforts: ['none', 'low', 'medium', 'high', 'xhigh', 'max'],
+  },
   'gpt-6-astra': {
-    aliases: ['astra', 'gpt-6', 'gpt6', 'gpt6-astra'],
+    aliases: ['astra', 'gpt6-astra'],
     contextWindow: 272000,
     supportedEfforts: ['low', 'medium', 'high', 'xhigh', 'max'],
   },
   'gpt-5.6-sol': {
-    aliases: ['sol', 'gpt-5.6', 'gpt5.6', 'gpt-5.6-codex'],
+    aliases: ['gpt-5.6', 'gpt5.6', 'gpt5.6-sol', 'gpt-5.6-codex'],
     contextWindow: 272000,
     supportedEfforts: ['none', 'low', 'medium', 'high', 'xhigh', 'max'],
   },
   'gpt-5.6-terra': {
-    aliases: ['terra'],
+    aliases: ['terra', 'gpt5.6-terra'],
     contextWindow: 272000,
     supportedEfforts: ['none', 'low', 'medium', 'high', 'xhigh', 'max'],
   },
   'gpt-5.6-luna': {
-    aliases: ['luna'],
+    aliases: ['gpt5.6-luna'],
     contextWindow: 272000,
     supportedEfforts: ['none', 'low', 'medium', 'high', 'xhigh', 'max'],
   },
@@ -67,14 +81,14 @@ const CODEX_BACKEND_MODELS = {
   },
 };
 
-const DEFAULT_BACKEND_MODEL = 'gpt-6-astra';
+const DEFAULT_BACKEND_MODEL = 'gpt-6-sol';
 
 // The single user-facing model the router exposes. The backend model behind it
 // comes from CODEX_MODEL, or from a `codex:<model>` spec.
 const SUPPORTED_MODELS = {
   codex: {
     modelName: 'codex',
-    friendlyName: 'OpenAI Codex (GPT-6 Astra)',
+    friendlyName: 'OpenAI Codex (GPT-6 Sol)',
     contextWindow: CODEX_BACKEND_MODELS[DEFAULT_BACKEND_MODEL].contextWindow,
     maxOutputTokens: 128000,
     supportsStreaming: true,
@@ -82,7 +96,7 @@ const SUPPORTED_MODELS = {
     supportsWebSearch: false, // Codex accesses files directly, not web
     timeout: 1800000, // 30 minutes
     description:
-      'OpenAI Codex agentic coding assistant with local file access and tool execution (GPT-6 Astra by default; pick another backend with codex:<model> or CODEX_MODEL)',
+      'OpenAI Codex agentic coding assistant with local file access and tool execution (GPT-6 Sol by default; pick another backend with codex:<model> or CODEX_MODEL)',
     aliases: [],
   },
 };
@@ -266,7 +280,7 @@ export function getBackendModelConfig(name) {
 /**
  * Resolve the requested model spec to the backend slug passed to the CLI.
  *
- * `codex` uses CODEX_MODEL (default gpt-6-astra); `codex:<model>` names a
+ * `codex` uses CODEX_MODEL (default gpt-6-sol); `codex:<model>` names a
  * backend directly, by slug or alias. Unknown names pass through verbatim so a
  * newly released model works before it is catalogued here — the CLI rejects
  * anything the backend does not know.

package/src/providers/copilot.js

@@ -35,11 +35,38 @@ const SUPPORTED_MODELS = {
   },
 
   // OpenAI models
-  // Bare `gpt-5.6` (and the legacy `gpt-5` shortcut) route to Sol, matching
-  // Copilot's own bare-alias behavior. Terra is the recommended balanced tier.
-  // `codex` and `gpt` point at the latest GPT tier (reachable only via the
-  // `copilot:` namespace — bare `codex` routes to the Codex provider and bare
-  // `gpt*` keyword-routes to OpenAI before Copilot's catalog is consulted).
+  // Bare `gpt-6` / `gpt-5.6` route to that generation's Sol, matching
+  // Copilot's own bare-alias behavior; `sol`/`luna` and the legacy `gpt-5`
+  // shortcut follow the current generation. `codex` and `gpt` point at the
+  // latest GPT tier (reachable only via the `copilot:` namespace — bare
+  // `codex` routes to the Codex provider and bare `gpt*` keyword-routes to
+  // OpenAI before Copilot's catalog is consulted).
+  'gpt-6-sol': {
+    modelName: 'gpt-6-sol',
+    friendlyName: 'GPT-6 Sol (via Copilot)',
+    contextWindow: 1050000,
+    maxOutputTokens: 32768,
+    supportsStreaming: true,
+    supportsImages: false,
+    supportsWebSearch: false,
+    supportsReasoningEffort: true,
+    timeout: 1800000,
+    description: 'OpenAI GPT-6 Sol via Copilot subscription',
+    aliases: ['gpt-6', 'gpt-5', 'gpt', 'codex', 'sol'],
+  },
+  'gpt-6-luna': {
+    modelName: 'gpt-6-luna',
+    friendlyName: 'GPT-6 Luna (via Copilot)',
+    contextWindow: 1050000,
+    maxOutputTokens: 32768,
+    supportsStreaming: true,
+    supportsImages: false,
+    supportsWebSearch: false,
+    supportsReasoningEffort: true,
+    timeout: 1800000,
+    description: 'OpenAI GPT-6 Luna via Copilot subscription',
+    aliases: ['luna'],
+  },
   'gpt-5.6-sol': {
     modelName: 'gpt-5.6-sol',
     friendlyName: 'GPT-5.6 Sol (via Copilot)',
@@ -51,7 +78,7 @@ const SUPPORTED_MODELS = {
     supportsReasoningEffort: true,
     timeout: 1800000,
     description: 'OpenAI GPT-5.6 Sol via Copilot subscription',
-    aliases: ['gpt-5.6', 'gpt-5', 'gpt', 'codex'],
+    aliases: ['gpt-5.6'],
   },
   'gpt-5.6-terra': {
     modelName: 'gpt-5.6-terra',
@@ -81,6 +108,19 @@ const SUPPORTED_MODELS = {
   },
 
   // Anthropic models
+  // Bare `opus` and `claude` follow the current Opus generation.
+  'claude-opus-5.5': {
+    modelName: 'claude-opus-5.5',
+    friendlyName: 'Claude Opus 5.5 (via Copilot)',
+    contextWindow: 200000,
+    maxOutputTokens: 32768,
+    supportsStreaming: true,
+    supportsImages: false,
+    supportsWebSearch: false,
+    timeout: 1800000,
+    description: 'Anthropic Claude Opus 5.5 via Copilot subscription',
+    aliases: ['opus', 'claude', 'claude-opus-5-5'],
+  },
   'claude-fable-5': {
     modelName: 'claude-fable-5',
     friendlyName: 'Claude Fable 5 (via Copilot)',
@@ -115,7 +155,7 @@ const SUPPORTED_MODELS = {
     supportsWebSearch: false,
     timeout: 1800000,
     description: 'Anthropic Claude Opus 5 via Copilot subscription',
-    aliases: ['opus', 'claude'],
+    aliases: [],
   },
   'claude-opus-4.8': {
     modelName: 'claude-opus-4.8',

package/src/providers/openai.js

@@ -10,21 +10,82 @@ import { debugLog, debugError } from '../utils/console.js';
 import { clampReasoningEffort } from '../utils/reasoningEffort.js';
 
 // Values each family accepts for reasoning effort, per the model pages at
-// developers.openai.com/api/docs/models. GPT-5.6 is the only family with
-// 'max'; the GPT-5.4 tier stops at 'xhigh'; the original GPT-5 minis kept
-// 'minimal' but never gained 'xhigh'; the o-series predates both ends of the
-// ladder; GPT-5.4 Pro starts at 'medium'. Models without a list are passed
-// the requested value unchanged, except uncatalogued GPT-5 Pro snapshots,
-// which are only known to accept 'high'.
-const GPT_56_EFFORT_TIERS = ['none', 'low', 'medium', 'high', 'xhigh', 'max'];
+// developers.openai.com/api/docs/models. The GPT-6 Sol/Luna tiers and the
+// GPT-5.6 family take the whole ladder; GPT-6 Astra has no 'none'; the
+// GPT-5.4 tier stops at 'xhigh'; the original GPT-5 minis kept 'minimal' but
+// never gained 'xhigh'; the o-series predates both ends of the ladder;
+// GPT-5.4 Pro starts at 'medium'. Models without a list are passed the
+// requested value unchanged, except uncatalogued GPT-5 Pro snapshots, which
+// are only known to accept 'high'.
+const FULL_EFFORT_TIERS = ['none', 'low', 'medium', 'high', 'xhigh', 'max'];
+const GPT_6_ASTRA_EFFORT_TIERS = ['low', 'medium', 'high', 'xhigh', 'max'];
 const GPT_54_EFFORT_TIERS = ['none', 'low', 'medium', 'high', 'xhigh'];
 const GPT_54_PRO_EFFORT_TIERS = ['medium', 'high', 'xhigh'];
 const GPT_5_EFFORT_TIERS = ['minimal', 'low', 'medium', 'high'];
 const O_SERIES_EFFORT_TIERS = ['low', 'medium', 'high'];
 const PRO_PASSTHROUGH_EFFORT_TIERS = ['high'];
 
-// Define supported models with their capabilities
+// Define supported models with their capabilities.
+// Bare tier names (sol, luna) and the bare generation (gpt-6, and the legacy
+// gpt-5 shortcut) follow the current generation; older tiers stay reachable by
+// their versioned names.
 const SUPPORTED_MODELS = {
+  'gpt-6-sol': {
+    modelName: 'gpt-6-sol',
+    friendlyName: 'OpenAI (GPT-6 Sol)',
+    contextWindow: 1050000,
+    maxOutputTokens: 128000,
+    supportsStreaming: true,
+    supportsImages: true,
+    supportsWebSearch: true,
+    supportsResponsesAPI: true,
+    supportedEfforts: FULL_EFFORT_TIERS,
+    timeout: 10800000, // 3 hours
+    description:
+      'Default GPT-6 model (1M context, 128K output) - Complex coding and agentic workflows at a fifth of the Astra price',
+    aliases: [
+      'gpt-6',
+      'gpt6',
+      'gpt 6',
+      'gpt-5',
+      'gpt5',
+      'gpt 5',
+      'sol',
+      'gpt6-sol',
+      'gpt-6sol',
+      'gpt 6 sol',
+    ],
+  },
+  'gpt-6-luna': {
+    modelName: 'gpt-6-luna',
+    friendlyName: 'OpenAI (GPT-6 Luna)',
+    contextWindow: 1050000,
+    maxOutputTokens: 128000,
+    supportsStreaming: true,
+    supportsImages: true,
+    supportsWebSearch: true,
+    supportsResponsesAPI: true,
+    supportedEfforts: FULL_EFFORT_TIERS,
+    timeout: 1800000, // 30 minutes
+    description:
+      'Most efficient GPT-6 (1M context, 128K output) - Focused, high-volume tasks',
+    aliases: ['luna', 'gpt6-luna', 'gpt-6luna', 'gpt 6 luna'],
+  },
+  'gpt-6-astra': {
+    modelName: 'gpt-6-astra',
+    friendlyName: 'OpenAI (GPT-6 Astra)',
+    contextWindow: 1050000,
+    maxOutputTokens: 128000,
+    supportsStreaming: true,
+    supportsImages: true,
+    supportsWebSearch: true,
+    supportsResponsesAPI: true,
+    supportedEfforts: GPT_6_ASTRA_EFFORT_TIERS,
+    timeout: 10800000, // 3 hours
+    description:
+      'Frontier GPT-6 flagship (1M context, 128K output) - Maximum intelligence for the hardest end-to-end work (EXPENSIVE: 5x Sol)',
+    aliases: ['astra', 'gpt6-astra', 'gpt-6astra', 'gpt 6 astra'],
+  },
   'gpt-5.6-sol': {
     modelName: 'gpt-5.6-sol',
     friendlyName: 'OpenAI (GPT-5.6 Sol)',
@@ -34,18 +95,15 @@ const SUPPORTED_MODELS = {
     supportsImages: true,
     supportsWebSearch: true,
     supportsResponsesAPI: true,
-    supportedEfforts: GPT_56_EFFORT_TIERS,
+    supportedEfforts: FULL_EFFORT_TIERS,
     timeout: 10800000, // 3 hours
     description:
-      'Flagship GPT-5.6 model (1M context, 128K output) - Frontier reasoning, coding, agentic workflows. Most token-efficient flagship',
+      'Previous flagship GPT-5.6 (1M context, 128K output) - Frontier reasoning, coding, agentic workflows',
     aliases: [
       'gpt-5.6',
       'gpt5.6',
       'gpt 5.6',
-      'gpt-5',
-      'gpt5',
-      'gpt 5',
-      'sol',
+      'gpt5.6-sol',
       'gpt-5.6sol',
       'gpt 5.6 sol',
     ],
@@ -59,7 +117,7 @@ const SUPPORTED_MODELS = {
     supportsImages: true,
     supportsWebSearch: true,
     supportsResponsesAPI: true,
-    supportedEfforts: GPT_56_EFFORT_TIERS,
+    supportedEfforts: FULL_EFFORT_TIERS,
     timeout: 5400000, // 90 minutes
     description:
       'Lower-cost GPT-5.6 (400K context, 128K output) - Performance competitive with GPT-5.5 at half the flagship price',
@@ -74,11 +132,11 @@ const SUPPORTED_MODELS = {
     supportsImages: true,
     supportsWebSearch: true,
     supportsResponsesAPI: true,
-    supportedEfforts: GPT_56_EFFORT_TIERS,
+    supportedEfforts: FULL_EFFORT_TIERS,
     timeout: 1800000, // 30 minutes
     description:
       'Fastest, most affordable GPT-5.6 (400K context, 128K output) - High-volume, latency-sensitive workloads',
-    aliases: ['gpt5.6-luna', 'gpt-5.6luna', 'gpt 5.6 luna', 'luna'],
+    aliases: ['gpt5.6-luna', 'gpt-5.6luna', 'gpt 5.6 luna'],
   },
   'gpt-5.4': {
     modelName: 'gpt-5.4',
@@ -342,14 +400,18 @@ function acceptsReasoningEffort(resolvedModel, modelConfig) {
   if (modelConfig.supportedEfforts) {
     return true;
   }
-  return resolvedModel.startsWith('o3') || resolvedModel.startsWith('gpt-5');
+  return (
+    resolvedModel.startsWith('o3') ||
+    resolvedModel.startsWith('gpt-5') ||
+    resolvedModel.startsWith('gpt-6')
+  );
 }
 
 /**
  * Resolve the reasoning effort actually sent to the API. Catalogued models
  * clamp onto their declared tiers. Pass-through IDs are matched by family
- * where the tiers are known (GPT-5 Pro snapshots, GPT-5.6 snapshots) and
- * otherwise keep the requested value.
+ * where the tiers are known (GPT-5 Pro snapshots, GPT-6 Sol/Luna and GPT-5.6
+ * snapshots) and otherwise keep the requested value.
  */
 function resolveReasoningEffort(resolvedModel, modelConfig, reasoningEffort) {
   if (modelConfig.supportedEfforts) {
@@ -358,8 +420,15 @@ function resolveReasoningEffort(resolvedModel, modelConfig, reasoningEffort) {
   if (resolvedModel.endsWith('-pro') && resolvedModel.startsWith('gpt-5')) {
     return clampReasoningEffort(reasoningEffort, PRO_PASSTHROUGH_EFFORT_TIERS);
   }
-  if (resolvedModel.startsWith('gpt-5.6')) {
-    return clampReasoningEffort(reasoningEffort, GPT_56_EFFORT_TIERS);
+  if (resolvedModel.startsWith('gpt-6-astra')) {
+    return clampReasoningEffort(reasoningEffort, GPT_6_ASTRA_EFFORT_TIERS);
+  }
+  if (
+    resolvedModel.startsWith('gpt-6-sol') ||
+    resolvedModel.startsWith('gpt-6-luna') ||
+    resolvedModel.startsWith('gpt-5.6')
+  ) {
+    return clampReasoningEffort(reasoningEffort, FULL_EFFORT_TIERS);
   }
   return reasoningEffort;
 }
@@ -487,7 +556,7 @@ export const openaiProvider = {
    */
   async invoke(messages, options = {}) {
     const {
-      model = 'gpt-5.6',
+      model = 'gpt-6',
       maxTokens = null,
       stream = false,
       reasoning_effort = 'medium',

package/src/utils/modelRouting.js

@@ -38,16 +38,16 @@ export function getDefaultModelForProvider(providerName) {
     'gemini-cli': 'gemini',
     claude: 'claude',
     copilot: 'copilot',
-    openai: 'gpt-5.6',
+    openai: 'gpt-6',
     xai: 'grok-4.5',
     google: 'gemini-pro',
-    anthropic: 'claude-sonnet-4-20250514',
+    anthropic: 'claude-opus-5-5',
     mistral: 'mistral-medium-3-5',
     deepseek: 'deepseek-v4-pro',
     openrouter: 'z-ai/glm-5.2',
   };
 
-  return defaults[providerName] || 'gpt-5.6';
+  return defaults[providerName] || 'gpt-6';
 }
 
 /**