@codexapi/codexclaude 2.0.21 → 2.0.23

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (5) hide show
  1. package/CUSTOMIZATIONS.md +11 -11
  2. package/README.md +8 -8
  3. package/bin/codexclaude.js +32 -21
  4. package/codexclaude.md +40 -29
  5. package/package.json +2 -2
package/CUSTOMIZATIONS.md CHANGED
@@ -5,9 +5,9 @@
5
5
  ## Runtime
6
6
 
7
7
  - Customer-facing app name: `Magnetapi Client`
8
- - Local launcher version: `2.0.21`
9
- - Bundled official runtime dependency: `@openai/codex@0.144.1`
10
- - Required install command: `npm install -g https://codexapi.pro/downloads/codexapi-codexclaude-2.0.21.tgz`
8
+ - Local launcher version: `2.0.23`
9
+ - Bundled official runtime dependency: `@openai/codex@0.144.3`
10
+ - Required install command: `npm install -g https://codexapi.pro/downloads/codexapi-codexclaude-2.0.23.tgz`
11
11
  - Bundled runtime version is managed internally and must not be used for customer update prompts.
12
12
  - Launcher update checks may call the CodexAPI.pro version endpoint and `@codexapi/codexclaude` npm metadata only.
13
13
  - The launcher must never check `@openai/codex` npm metadata or OpenAI GitHub releases for update notices.
@@ -24,7 +24,7 @@
24
24
  - Passes unknown commands directly to the bundled runtime after config is applied.
25
25
  - Provides `codexclaude changelog` using the managed version endpoint.
26
26
  - Does not expose separate `codexclaude status` or `codexclaude usage` commands; wallet information remains a compact pre-launch line.
27
- - Version `2.0.21` is forced by the backend minimum version and older clients must show an update-required message.
27
+ - Version `2.0.23` is forced by the backend minimum version and older clients must show an update-required message.
28
28
 
29
29
  ## Provider Config
30
30
 
@@ -41,9 +41,9 @@ The launcher writes a model provider named `codexapi`:
41
41
 
42
42
  ## Runtime Defaults
43
43
 
44
- - Default model: `gpt-5.6-sol` displayed as `GPT-5.6 SOL`
45
- - Supported launcher choices: `gpt-5.6-sol-max`, `gpt-5.6-sol-ultra`, `gpt-5.6-sol`, `gpt-5.5`, `claude-opus-4-8`, `claude-sonnet-5`, and `claude-fable-5`
46
- - Reasoning effort displayed locally: model-specific (`low`, `max`, `ultra`, or `xhigh` fallback)
44
+ - Default model: `gpt-5.6` displayed as `GPT-5.6`
45
+ - Supported launcher choices: `gpt-5.6`, `gpt-5.6-sol-max`, `gpt-5.6-sol-ultra`, `gpt-5.6-sol`, `gpt-5.5`, `claude-opus-4-8`, `claude-sonnet-5`, and `claude-fable-5`
46
+ - Default reasoning effort displayed locally: `xhigh`; SOL Max and SOL Ultra retain their explicit premium labels.
47
47
  - Reasoning hint sent in launcher env: model-specific
48
48
  - Approval policy: `never`
49
49
  - Sandbox mode: `danger-full-access`
@@ -51,8 +51,8 @@ The launcher writes a model provider named `codexapi`:
51
51
  - Local model catalog: `~/.codexclaude/model_catalog.json`; this keeps the native `/model` picker limited to the supported choices and prevents fallback metadata warnings for GPT-5.6 SOL, SOL Max, SOL Ultra, Opus 4.8, Sonnet 5, and Fable 5.
52
52
  - Service tier: omitted by default; `/fast` remains user-controlled and billed server-side when active.
53
53
  - TUI status line: `model-with-reasoning`, `fast-mode`, `current-dir`
54
- - Fable 5 stays visible to the user while the backend applies its custom routed premium path and billing multiplier.
55
- - GPT-5.6 SOL Max bills at 2x normal SOL usage. GPT-5.6 SOL Ultra bills at 2.5x normal SOL usage.
54
+ - Fable 5 stays visible to the user while the backend applies its server-managed route and billing policy.
55
+ - Premium model billing remains server-managed and is not rendered in the customer model picker or model descriptions.
56
56
 
57
57
  ## Wallet Display
58
58
 
@@ -61,7 +61,7 @@ Before launching the runtime, the launcher calls `/api/v1/wallet/status`, then `
61
61
  Expected status line:
62
62
 
63
63
  ```text
64
- Magnetapi Client v2.0.21 · gpt-5.6-sol low · ~/project · Wallet $20.00
64
+ Magnetapi Client v2.0.23 · gpt-5.6 xhigh · ~/project · Wallet $20.00
65
65
  ```
66
66
 
67
67
  ## Upgrade Checklist
@@ -72,7 +72,7 @@ When updating the bundled runtime:
72
72
  2. Update `@openai/codex` in `package.json` and `package-lock.json`.
73
73
  3. Bump the launcher package version.
74
74
  4. Keep `APP_DISPLAY_NAME = "Magnetapi Client"`.
75
- 5. Keep `DEFAULT_MODEL_ID = "gpt-5.6-sol"`.
75
+ 5. Keep `DEFAULT_MODEL_ID = "gpt-5.6"`.
76
76
  6. Keep `reasoningEffortForModel()` and model-specific visible reasoning.
77
77
  7. Keep Full Access defaults: `approval_policy = "never"` and `sandbox_mode = "danger-full-access"`.
78
78
  8. Keep `check_for_update_on_startup = false`.
package/README.md CHANGED
@@ -1,13 +1,13 @@
1
1
  # codexclaude
2
2
 
3
- `codexclaude` installs the Magnetapi Client for customers who want one coding CLI that can use GPT-5.6 SOL, GPT-5.6 SOL Max, GPT-5.6 SOL Ultra, ChatGPT-5.5, Opus 4.8, Sonnet 5, or Fable 5 against the same Token Wallet.
3
+ `codexclaude` installs the Magnetapi Client for customers who want one coding CLI that can use GPT-5.6, GPT-5.6 SOL, GPT-5.6 SOL Max, GPT-5.6 SOL Ultra, ChatGPT-5.5, Opus 4.8, Sonnet 5, or Fable 5 against the same Token Wallet.
4
4
 
5
- Version `2.0.21` is a required major sync release. It bundles official Codex runtime dependency `@openai/codex@0.144.1`, launches the platform-native runtime binary directly, disables upstream OpenAI update prompts, and applies account, wallet, model, Full Access, GPT-5.6 SOL default, and model-specific visible reasoning before each coding session starts.
5
+ Version `2.0.23` is a required client sync release. It bundles official Codex runtime dependency `@openai/codex@0.144.3`, launches the platform-native runtime binary directly, disables upstream OpenAI update prompts, and applies account, wallet, model, Full Access, GPT-5.6/xhigh defaults, and managed model metadata before each coding session starts. Model menus no longer display wallet multiplier notices; billing remains managed by the server.
6
6
 
7
7
  ## Install Or Update
8
8
 
9
9
  ```bash
10
- npm install -g https://codexapi.pro/downloads/codexapi-codexclaude-2.0.21.tgz
10
+ npm install -g https://codexapi.pro/downloads/codexapi-codexclaude-2.0.23.tgz
11
11
  ```
12
12
 
13
13
  This does not delete local session history. The update replaces the `codexclaude` launcher package, refreshes the bundled runtime dependency, and rewrites the managed provider config the next time `codexclaude setup`, `codexclaude login`, or `codexclaude start` runs.
@@ -16,7 +16,7 @@ This does not delete local session history. The update replaces the `codexclaude
16
16
 
17
17
  The launcher checks CodexAPI.pro's managed version endpoint and the `@codexapi/codexclaude` npm package metadata only. It must never check `@openai/codex` npm metadata or OpenAI GitHub releases for customer update prompts.
18
18
 
19
- Version `2.0.21` is currently enforced as the minimum supported version, so older launchers show a required update message with the exact hosted tarball command above.
19
+ Version `2.0.23` is currently enforced as the minimum supported version, so older launchers show a required update message with the exact hosted tarball command above.
20
20
 
21
21
  ## First Login
22
22
 
@@ -24,7 +24,7 @@ Version `2.0.21` is currently enforced as the minimum supported version, so olde
24
24
  codexclaude login
25
25
  ```
26
26
 
27
- Enter the Dashboard Username or Email Address and password. The launcher can log in to CodexAPI.pro, ChatPlus.chat, MyAPI.world, or MagnetAPI.org, then saves the correct System Token locally, writes `~/.codex/config.toml`, sets Full Access defaults, and lets the customer choose the coding model. If the customer presses Enter at the model prompt, GPT-5.6 SOL is selected by default.
27
+ Enter the Dashboard Username or Email Address and password. The launcher can log in to CodexAPI.pro, ChatPlus.chat, MyAPI.world, or MagnetAPI.org, then saves the correct System Token locally, writes `~/.codex/config.toml`, sets Full Access defaults, and lets the customer choose the coding model. If the customer presses Enter at the model prompt, GPT-5.6 with xhigh reasoning is selected by default.
28
28
 
29
29
  ## Start Coding
30
30
 
@@ -44,7 +44,7 @@ Switch model:
44
44
  codexclaude model
45
45
  ```
46
46
 
47
- Available model labels are `GPT-5.6 SOL Max`, `GPT-5.6 SOL Ultra`, `GPT-5.6 SOL`, `ChatGPT-5.5`, `Opus 4.8`, `Sonnet 5`, and `Fable 5`. SOL Max bills at 2x normal SOL usage. SOL Ultra bills at 2.5x normal SOL usage. All models use a Responses-compatible endpoint and the same customer Token Wallet.
47
+ Available model labels are `GPT-5.6`, `GPT-5.6 SOL Max`, `GPT-5.6 SOL Ultra`, `GPT-5.6 SOL`, `ChatGPT-5.5`, `Opus 4.8`, `Sonnet 5`, and `Fable 5`. All models use a Responses-compatible endpoint and the same customer Token Wallet.
48
48
 
49
49
  Show current launcher changes:
50
50
 
@@ -57,13 +57,13 @@ codexclaude changelog
57
57
  Before launching the coding runtime, the app prints a compact status line like:
58
58
 
59
59
  ```text
60
- Magnetapi Client v2.0.21 · gpt-5.6-sol low · ~/project · Wallet $20.00
60
+ Magnetapi Client v2.0.23 · gpt-5.6 xhigh · ~/project · Wallet $20.00
61
61
  ```
62
62
 
63
63
  ## Preserved Customizations
64
64
 
65
65
  - Dashboard Username or Email Address plus password login for CodexAPI.pro, ChatPlus.chat, MyAPI.world, and MagnetAPI.org.
66
- - GPT-5.6 SOL Max, GPT-5.6 SOL Ultra, GPT-5.6 SOL, ChatGPT-5.5, Opus 4.8, Sonnet 5, or Fable 5 model selection, with GPT-5.6 SOL as the default.
66
+ - GPT-5.6, GPT-5.6 SOL Max, GPT-5.6 SOL Ultra, GPT-5.6 SOL, ChatGPT-5.5, Opus 4.8, Sonnet 5, or Fable 5 model selection, with GPT-5.6/xhigh as the default.
67
67
  - Service-specific Responses-compatible provider for the dashboard account that logged in.
68
68
  - Full Access defaults: approval policy `never` and sandbox mode `danger-full-access`.
69
69
  - Model-specific visible reasoning in the local UI while the API server keeps upstream runtime effort server-managed.
package/bin/codexclaude.js CHANGED
@@ -10,10 +10,10 @@ import { stdin as input, stdout as output } from "node:process";
10
10
 
11
11
  const require = createRequire(import.meta.url);
12
12
 
13
- const APP_VERSION = "2.0.21";
14
- const CODEX_VERSION = "0.144.1";
13
+ const APP_VERSION = "2.0.23";
14
+ const CODEX_VERSION = "0.144.3";
15
15
  const APP_DISPLAY_NAME = "Magnetapi Client";
16
- const DEFAULT_MODEL_ID = "gpt-5.6-sol";
16
+ const DEFAULT_MODEL_ID = "gpt-5.6";
17
17
  const DEFAULT_DISPLAY_REASONING_EFFORT = "xhigh";
18
18
  const DEFAULT_SERVER_REASONING_EFFORT = "xhigh";
19
19
  const API_ORIGIN = (process.env.CODEXCLAUDE_API_ORIGIN || "https://codexapi.pro").replace(/\/+$/, "");
@@ -74,47 +74,50 @@ const SUPPORTED_REASONING_LEVELS = [
74
74
  ];
75
75
 
76
76
  const MODELS = {
77
+ "gpt-5.6": {
78
+ id: "gpt-5.6",
79
+ label: "GPT-5.6",
80
+ aliases: ["1", "gpt", "gpt-5.6", "chatgpt", "chatgpt-5.6", "gpt56"],
81
+ reasoningEffort: "xhigh",
82
+ },
77
83
  "gpt-5.6-sol-max": {
78
84
  id: "gpt-5.6-sol-max",
79
85
  label: "GPT-5.6 SOL Max",
80
- aliases: ["1", "gpt-5.6-sol-max", "gpt-5.6-max", "gpt56-max", "sol-max", "max"],
81
- premiumLabel: "2x premium billing",
86
+ aliases: ["2", "gpt-5.6-sol-max", "gpt-5.6-max", "gpt56-max", "sol-max", "max"],
82
87
  reasoningEffort: "max",
83
88
  },
84
89
  "gpt-5.6-sol-ultra": {
85
90
  id: "gpt-5.6-sol-ultra",
86
91
  label: "GPT-5.6 SOL Ultra",
87
- aliases: ["2", "gpt-5.6-sol-ultra", "gpt-5.6-ultra", "gpt56-ultra", "sol-ultra", "ultra"],
88
- premiumLabel: "2.5x premium billing",
92
+ aliases: ["3", "gpt-5.6-sol-ultra", "gpt-5.6-ultra", "gpt56-ultra", "sol-ultra", "ultra"],
89
93
  reasoningEffort: "ultra",
90
94
  },
91
95
  "gpt-5.6-sol": {
92
96
  id: "gpt-5.6-sol",
93
97
  label: "GPT-5.6 SOL",
94
- aliases: ["3", "gpt-5.6-sol", "gpt-5.6", "gpt56", "sol"],
95
- reasoningEffort: "low",
98
+ aliases: ["4", "gpt-5.6-sol", "sol"],
99
+ reasoningEffort: "xhigh",
96
100
  },
97
101
  "gpt-5.5": {
98
102
  id: "gpt-5.5",
99
103
  label: "ChatGPT-5.5",
100
- aliases: ["4", "gpt", "gpt-5.5", "chatgpt", "chatgpt-5.5"],
104
+ aliases: ["5", "gpt-5.5", "chatgpt-5.5"],
101
105
  },
102
106
  "claude-opus-4-8": {
103
107
  id: "claude-opus-4-8",
104
108
  label: "Opus 4.8",
105
- aliases: ["5", "opus", "opus-4.8", "opus-4.7", "claude", "claude-opus-4-8", "claude-opus-4-7"],
109
+ aliases: ["6", "opus", "opus-4.8", "opus-4.7", "claude", "claude-opus-4-8", "claude-opus-4-7"],
106
110
  },
107
111
  "claude-sonnet-5": {
108
112
  id: "claude-sonnet-5",
109
113
  label: "Sonnet 5",
110
- aliases: ["6", "sonnet", "sonnet-5", "sonnet-4.6", "sonnet-4-6", "claude-sonnet-5", "claude-sonnet-4-6", "claude-sonnet-4.6"],
114
+ aliases: ["7", "sonnet", "sonnet-5", "sonnet-4.6", "sonnet-4-6", "claude-sonnet-5", "claude-sonnet-4-6", "claude-sonnet-4.6"],
111
115
  },
112
116
  "claude-fable-5": {
113
117
  id: "claude-fable-5",
114
118
  label: "Fable 5",
115
- aliases: ["7", "fable", "fable-5", "fabel", "fabel-5", "claude-fable-5", "claude-fabel-5"],
119
+ aliases: ["8", "fable", "fable-5", "fabel", "fabel-5", "claude-fable-5", "claude-fabel-5"],
116
120
  premium: true,
117
- premiumLabel: "2x premium billing",
118
121
  },
119
122
  };
120
123
 
@@ -171,7 +174,7 @@ ${APP_DISPLAY_NAME} ${APP_VERSION}
171
174
 
172
175
  Commands:
173
176
  codexclaude login Login with dashboard username/email and password
174
- codexclaude model Select ChatGPT-5.5, Opus 4.8, Sonnet 5, or Fable 5
177
+ codexclaude model Select GPT-5.6, SOL modes, Opus 4.8, Sonnet 5, or Fable 5
175
178
  codexclaude changelog Show the latest CodexAPI.pro launcher changes
176
179
  codexclaude setup Rewrite ~/.codex/config.toml with Full Access defaults
177
180
  codexclaude start Start Magnetapi Client with --search
@@ -371,8 +374,7 @@ async function askModel(rl, current) {
371
374
  const currentLabel = MODELS[current]?.label || MODELS[DEFAULT_MODEL_ID].label;
372
375
  console.log("\nChoose your coding model:");
373
376
  Object.values(MODELS).forEach((model, index) => {
374
- const premium = model.premiumLabel ? ` (${model.premiumLabel})` : model.premium ? " (2x premium billing)" : "";
375
- console.log(` ${index + 1}. ${model.label}${premium}`);
377
+ console.log(` ${index + 1}. ${model.label}`);
376
378
  });
377
379
  const answer = (await rl.question(`Model [${currentLabel}]: `)).trim().toLowerCase();
378
380
  if (!answer) {
@@ -699,11 +701,14 @@ function modelCatalogEntry(model, index) {
699
701
  }
700
702
 
701
703
  function modelCatalogDescription(modelId) {
704
+ if (modelId === "gpt-5.6") {
705
+ return "GPT-5.6 coding route with local coding tools and wallet billing.";
706
+ }
702
707
  if (modelId === "gpt-5.6-sol-max") {
703
- return "GPT-5.6 SOL Max premium coding route with local coding tools and 2x wallet billing.";
708
+ return "GPT-5.6 SOL Max premium coding route with local coding tools and wallet billing.";
704
709
  }
705
710
  if (modelId === "gpt-5.6-sol-ultra") {
706
- return "GPT-5.6 SOL Ultra premium coding route with local coding tools and 2.5x wallet billing.";
711
+ return "GPT-5.6 SOL Ultra premium coding route with local coding tools and wallet billing.";
707
712
  }
708
713
  if (modelId === "gpt-5.6-sol") {
709
714
  return "GPT-5.6 SOL native coding route with local coding tools and wallet billing.";
@@ -923,6 +928,9 @@ function formatStatusLine(config, wallet, cwd) {
923
928
  }
924
929
 
925
930
  function modelStatusName(model) {
931
+ if (model === "gpt-5.6") {
932
+ return "gpt-5.6";
933
+ }
926
934
  if (model === "gpt-5.6-sol-max") {
927
935
  return "gpt-5.6-sol max";
928
936
  }
@@ -941,7 +949,7 @@ function modelStatusName(model) {
941
949
  if (model === "claude-fable-5") {
942
950
  return "fable-5";
943
951
  }
944
- return "gpt-5.5";
952
+ return "gpt-5.6";
945
953
  }
946
954
 
947
955
  function reasoningEffortForModel(model) {
@@ -968,9 +976,12 @@ function normalizeConfig(config) {
968
976
  if (["gpt-5.6-sol-ultra", "gpt-5.6-ultra", "gpt56-ultra", "sol-ultra", "ultra"].includes(normalized.model)) {
969
977
  normalized.model = "gpt-5.6-sol-ultra";
970
978
  }
971
- if (["gpt-5-6-sol", "gpt56", "sol"].includes(normalized.model)) {
979
+ if (["gpt-5-6-sol", "sol"].includes(normalized.model)) {
972
980
  normalized.model = "gpt-5.6-sol";
973
981
  }
982
+ if (["gpt-5-6", "gpt56", "chatgpt-5.6", "chatgpt-5-6"].includes(normalized.model)) {
983
+ normalized.model = "gpt-5.6";
984
+ }
974
985
  if (
975
986
  [
976
987
  "deepseek-v4-pro",
package/codexclaude.md CHANGED
@@ -5,14 +5,14 @@ This file is the durable handoff for the `@codexapi/codexclaude` package in `/ro
5
5
  ## Current Release
6
6
 
7
7
  - Package name: `@codexapi/codexclaude`
8
- - Launcher version: `2.0.21`
8
+ - Launcher version: `2.0.23`
9
9
  - Customer-facing app name: `Magnetapi Client`
10
- - Bundled runtime dependency: `@openai/codex@0.144.1`
11
- - Required install command: `npm install -g https://codexapi.pro/downloads/codexapi-codexclaude-2.0.21.tgz`
12
- - Backend minimum version: `2.0.21`
13
- - Default visible model: `GPT-5.6 SOL`
14
- - Default model slug: `gpt-5.6-sol`
15
- - Visible reasoning effort: model-specific (`low` for base SOL, `max` for SOL Max, `ultra` for SOL Ultra, `xhigh` fallback for the other managed models)
10
+ - Bundled runtime dependency: `@openai/codex@0.144.3`
11
+ - Required install command: `npm install -g https://codexapi.pro/downloads/codexapi-codexclaude-2.0.23.tgz`
12
+ - Backend minimum version: `2.0.23`
13
+ - Default visible model: `GPT-5.6`
14
+ - Default model slug: `gpt-5.6`
15
+ - Default visible reasoning effort: `xhigh`; premium SOL Max and SOL Ultra retain their explicit `max` and `ultra` labels.
16
16
  - Local permissions: Full Access, implemented as `approval_policy = "never"` and `sandbox_mode = "danger-full-access"`
17
17
  - Local launcher state: `~/.codexclaude/config.json`
18
18
  - Generated runtime config: `~/.codex/config.toml`
@@ -24,28 +24,29 @@ This file is the durable handoff for the `@codexapi/codexclaude` package in `/ro
24
24
  The launcher must present itself as:
25
25
 
26
26
  ```text
27
- Magnetapi Client v2.0.21
27
+ Magnetapi Client v2.0.23
28
28
  ```
29
29
 
30
30
  Normal pre-launch status should look like:
31
31
 
32
32
  ```text
33
- Magnetapi Client v2.0.21 · gpt-5.6-sol low · ~/project · Wallet $20.00
33
+ Magnetapi Client v2.0.23 · gpt-5.6 xhigh · ~/project · Wallet $20.00
34
34
  ```
35
35
 
36
36
  Do not show OpenAI update notices, OpenAI npm install commands, OpenAI release links, or the upstream runtime version in launcher-owned output. The package still depends on the official runtime internally, but the customer update surface is ours.
37
37
 
38
- The launcher command `codexclaude model` must list these customer-visible models, with SOL Max and Ultra as choices 1 and 2:
38
+ The launcher command `codexclaude model` must list these customer-visible models:
39
39
 
40
- - `gpt-5.6-sol-max` displayed as `GPT-5.6 SOL Max` with 2x premium billing
41
- - `gpt-5.6-sol-ultra` displayed as `GPT-5.6 SOL Ultra` with 2.5x premium billing
40
+ - `gpt-5.6` displayed as `GPT-5.6`, first and selected by default
41
+ - `gpt-5.6-sol-max` displayed as `GPT-5.6 SOL Max`
42
+ - `gpt-5.6-sol-ultra` displayed as `GPT-5.6 SOL Ultra`
42
43
  - `gpt-5.6-sol` displayed as `GPT-5.6 SOL`
43
44
  - `gpt-5.5` displayed as `ChatGPT-5.5`
44
45
  - `claude-opus-4-8` displayed as `Opus 4.8`
45
46
  - `claude-sonnet-5` displayed as `Sonnet 5`
46
47
  - `claude-fable-5` displayed as `Fable 5`
47
48
 
48
- GPT-5.6 SOL must be the default when a new config is created or a legacy config contains an unsupported Gemini/Deepseek value. `gpt-5.6-sol` is now a native upstream model on the signed-in Codex API servers and is pinned to server-managed low reasoning. The selected model slug must stay in `~/.codex/config.toml`; do not rewrite customer-facing model selections locally just because the backend may route internally. `gpt-5.6-sol-max` and `gpt-5.6-sol-ultra` still run through the native SOL route but must remain distinct for customer choice, usage display, and wallet billing.
49
+ GPT-5.6 must be the default when a new config is created or a legacy config contains an unsupported Gemini/Deepseek value. It is displayed with `xhigh` reasoning while the API servers execute Codex-family requests through the managed GPT-5.6 Terra/xhigh signed-in target. The selected model slug must stay in `~/.codex/config.toml`; do not rewrite customer-facing model selections locally. `gpt-5.6-sol-max` and `gpt-5.6-sol-ultra` remain distinct for customer choice, usage display, and wallet billing.
49
50
 
50
51
  ## Update Policy
51
52
 
@@ -54,7 +55,7 @@ GPT-5.6 SOL must be the default when a new config is created or a legacy config
54
55
  - `https://codexapi.pro/api/v1/codexclaude/version`
55
56
  - `https://registry.npmjs.org/@codexapi%2Fcodexclaude/latest`
56
57
 
57
- It must not check `@openai/codex` npm metadata, OpenAI GitHub releases, or any upstream OpenAI update endpoint. Version `2.0.21` is a forced release: the backend `minimumVersion` equals `2.0.21`, and older clients must show an update-required message with the hosted tarball command.
58
+ It must not check `@openai/codex` npm metadata, OpenAI GitHub releases, or any upstream OpenAI update endpoint. Version `2.0.23` is a forced release: the backend `minimumVersion` equals `2.0.23`, and older clients must show an update-required message with the hosted tarball command.
58
59
 
59
60
  The launcher writes `check_for_update_on_startup = false` and removes `~/.codex/version.json` so the bundled runtime does not display stale upstream update prompts. The launcher also resolves the platform-native runtime binary directly from the optional package (`@openai/codex-linux-x64`, `@openai/codex-darwin-arm64`, and so on). Do not fall back to the upstream npm wrapper, because that wrapper can re-enable upstream managed update behavior.
60
61
 
@@ -65,7 +66,7 @@ The optional platform-native runtime packages can still contain upstream README
65
66
  `bin/codexclaude.js` must rewrite `~/.codex/config.toml` before every coding session. Preserve this top-level shape:
66
67
 
67
68
  ```toml
68
- model = "gpt-5.6-sol"
69
+ model = "gpt-5.6"
69
70
  model_catalog_json = "<absolute path to ~/.codexclaude/model_catalog.json>"
70
71
  preferred_auth_method = "apikey"
71
72
  model_provider = "codexapi"
@@ -82,7 +83,7 @@ base_url = "<service Responses URL>"
82
83
  experimental_bearer_token = "<customer System Token/API username>"
83
84
  wire_api = "responses"
84
85
  supports_websockets = false
85
- http_headers = { "X-CodexAPI-Client" = "codexclaude", "X-CodexAPI-Client-Version" = "2.0.21", "X-CodexAPI-Selected-Model" = "<selected model>" }
86
+ http_headers = { "X-CodexAPI-Client" = "codexclaude", "X-CodexAPI-Client-Version" = "2.0.23", "X-CodexAPI-Selected-Model" = "<selected model>" }
86
87
 
87
88
  [tui]
88
89
  status_line = ["model-with-reasoning", "fast-mode", "current-dir"]
@@ -90,7 +91,7 @@ status_line_use_colors = true
90
91
  terminal_title = ["project-name", "model-with-reasoning", "run-state"]
91
92
  ```
92
93
 
93
- The generated model catalog must contain full metadata for every supported slug, including `gpt-5.6-sol-max` and `gpt-5.6-sol-ultra`. This prevents fallback metadata warnings such as:
94
+ The generated model catalog must contain full metadata for every supported slug, including `gpt-5.6`, `gpt-5.6-sol-max`, and `gpt-5.6-sol-ultra`. This prevents fallback metadata warnings such as:
94
95
 
95
96
  ```text
96
97
  Model metadata for `claude-opus-4-8` not found. Defaulting to fallback metadata; this can degrade performance and cause issues.
@@ -114,7 +115,7 @@ The saved `username` is the System Token/API username returned by the service an
114
115
 
115
116
  The launcher is not the source of truth for upstream runtime routing. Admin Manage Models controls execution routing server-side. The launcher must preserve the user-selected model slug and pass it through the Responses request so billing, logs, and the dashboard can record what the user chose.
116
117
 
117
- CodexAPI.pro and ChatPlus Claude-family models use the CodexAPI.pro Claude-compatible Responses endpoint. MyAPI.world and MagnetAPI.org accounts keep their own API origins. Fable 5 remains visible as `claude-fable-5` while the backend applies the premium model behavior and billing. SOL Max bills at 2x normal SOL usage, and SOL Ultra bills at 2.5x normal SOL usage.
118
+ CodexAPI.pro and ChatPlus Claude-family models use the CodexAPI.pro Claude-compatible Responses endpoint. MyAPI.world and MagnetAPI.org accounts keep their own API origins. Fable 5 remains visible as `claude-fable-5` while the backend applies the premium model behavior and billing. Premium billing stays server-managed and is not added to assistant output or customer model descriptions.
118
119
 
119
120
  ## Upgrade Checklist
120
121
 
@@ -125,7 +126,7 @@ When updating the bundled runtime dependency:
125
126
  3. Bump `package.json` and `package-lock.json` to a new `@codexapi/codexclaude` version.
126
127
  4. Update `APP_VERSION` and `CODEX_VERSION` in `bin/codexclaude.js`.
127
128
  5. Preserve `APP_DISPLAY_NAME = "Magnetapi Client"`.
128
- 6. Preserve `DEFAULT_MODEL_ID = "gpt-5.6-sol"`.
129
+ 6. Preserve `DEFAULT_MODEL_ID = "gpt-5.6"`.
129
130
  7. Preserve `DISPLAY_REASONING_EFFORT = "xhigh"` and `SERVER_REASONING_EFFORT = "xhigh"` unless operations explicitly changes the launcher-visible contract.
130
131
  8. Preserve Full Access defaults and the local model catalog.
131
132
  9. Preserve direct platform-native binary launch.
@@ -152,7 +153,7 @@ tmp_home="$(mktemp -d)"
152
153
  mkdir -p "$tmp_home/.codexclaude"
153
154
  cat > "$tmp_home/.codexclaude/config.json" <<'JSON'
154
155
  {
155
- "version": "2.0.21",
156
+ "version": "2.0.23",
156
157
  "serviceId": "codexapi",
157
158
  "serviceLabel": "CodexAPI.pro",
158
159
  "apiOrigin": "https://codexapi.pro",
@@ -167,7 +168,7 @@ HOME="$tmp_home" CODEXCLAUDE_DISABLE_UPDATE_CHECK=1 CODEXCLAUDE_QUIET_STATUS=1 n
167
168
 
168
169
  Verify the generated config contains:
169
170
 
170
- - `model = "gpt-5.6-sol"`
171
+ - `model = "gpt-5.6"`
171
172
  - `model_reasoning_effort = "xhigh"`
172
173
  - `check_for_update_on_startup = false`
173
174
  - `approval_policy = "never"`
@@ -176,6 +177,10 @@ Verify the generated config contains:
176
177
 
177
178
  Verify the generated model catalog contains only:
178
179
 
180
+ - `gpt-5.6`
181
+ - `gpt-5.6-sol-max`
182
+ - `gpt-5.6-sol-ultra`
183
+ - `gpt-5.6-sol`
179
184
  - `gpt-5.5`
180
185
  - `claude-opus-4-8`
181
186
  - `claude-sonnet-5`
@@ -185,21 +190,27 @@ After packaging, copy the tarball to the public downloads directories, restart t
185
190
 
186
191
  ```bash
187
192
  curl -fsS "https://codexapi.pro/api/v1/codexclaude/version?current=2.0.17"
188
- curl -fsSI "https://codexapi.pro/downloads/codexapi-codexclaude-2.0.21.tgz"
193
+ curl -fsSI "https://codexapi.pro/downloads/codexapi-codexclaude-2.0.23.tgz"
189
194
  ```
190
195
 
191
- The version endpoint must report `latestVersion = 2.0.21`, `minimumVersion = 2.0.21`, and `updateRequired = true` for old clients.
196
+ The version endpoint must report `latestVersion = 2.0.23`, `minimumVersion = 2.0.23`, and `updateRequired = true` for old clients.
192
197
 
193
- ## Version 2.0.21 Changes
198
+ ## Version 2.0.23 Changes
194
199
 
195
- - Updated bundled runtime dependency to `@openai/codex@0.144.1`.
196
- - Forced all older codexclaude clients to update by setting backend minimum version to `2.0.21`.
200
+ - Removed wallet multiplier notices from the customer model picker and model descriptions.
201
+ - Kept premium billing and route selection server-managed without altering assistant output.
202
+ - Required older clients to synchronize so the removed notices cannot persist locally.
203
+
204
+ ## Version 2.0.22 Changes
205
+
206
+ - Updated bundled runtime dependency to `@openai/codex@0.144.3`.
207
+ - Forced all older codexclaude clients to update by setting backend minimum version to `2.0.22`.
197
208
  - Changed launcher-owned identity to `Magnetapi Client`.
198
- - Made GPT-5.6 SOL a native signed-in Codex model route with low server reasoning.
209
+ - Made GPT-5.6 with xhigh reasoning the default customer-visible model.
199
210
  - Added GPT-5.6 SOL Max and GPT-5.6 SOL Ultra as explicit premium model choices.
200
- - Changed visible reasoning to model-specific values: low, max, ultra, or xhigh fallback.
211
+ - Preserved xhigh as the standard visible reasoning while retaining premium Max and Ultra labels.
201
212
  - Preserved Full Access defaults.
202
213
  - Disabled bundled runtime update checks through generated config and stale update-cache removal.
203
214
  - Removed launcher fallback to the upstream npm wrapper so OpenAI update prompts are not emitted by codexclaude.
204
215
  - Kept update monitoring limited to CodexAPI.pro version metadata and the `@codexapi/codexclaude` npm package.
205
- - Updated setup scripts to install the exact hosted `2.0.21` tarball.
216
+ - Updated setup scripts to install the exact hosted `2.0.22` tarball.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@codexapi/codexclaude",
3
- "version": "2.0.21",
3
+ "version": "2.0.23",
4
4
  "description": "Magnetapi Client launcher for ChatGPT-5.5, Opus 4.8, Sonnet 5, and Fable 5.",
5
5
  "license": "Apache-2.0",
6
6
  "type": "module",
@@ -17,7 +17,7 @@
17
17
  "node": ">=20"
18
18
  },
19
19
  "dependencies": {
20
- "@openai/codex": "0.144.1"
20
+ "@openai/codex": "0.144.3"
21
21
  },
22
22
  "publishConfig": {
23
23
  "access": "public"