consult-llm-mcp 2.4.2 → 2.5.0

package/README.md

@@ -456,9 +456,9 @@ See the "Using web mode..." example above for a concrete transcript.
   mode)
 - `DEEPSEEK_API_KEY` - Your DeepSeek API key (required for DeepSeek models)
 - `CONSULT_LLM_DEFAULT_MODEL` - Override the default model (optional)
-  - Options: `gpt-5.2` (default), `gemini-2.5-pro`, `gemini-3-pro-preview`,
-    `gemini-3.1-pro-preview`, `deepseek-reasoner`, `gpt-5.3-codex`,
-    `gpt-5.2-codex`
+  - Options: `gpt-5.2` (default), `gpt-5.4`, `gemini-2.5-pro`,
+    `gemini-3-pro-preview`, `gemini-3.1-pro-preview`, `deepseek-reasoner`,
+    `gpt-5.3-codex`, `gpt-5.2-codex`
 - `GEMINI_BACKEND` - Backend for Gemini models (optional)
   - Options: `api` (default), `gemini-cli`, `cursor-cli`
 - `OPENAI_BACKEND` - Backend for OpenAI models (optional)
@@ -555,9 +555,9 @@ models complex questions.
   - All files are added as context with file paths and code blocks
 
 - **model** (optional): LLM model to use
-  - Options: `gpt-5.2` (default), `gemini-2.5-pro`, `gemini-3-pro-preview`,
-    `gemini-3.1-pro-preview`, `deepseek-reasoner`, `gpt-5.3-codex`,
-    `gpt-5.2-codex`
+  - Options: `gpt-5.2` (default), `gpt-5.4`, `gemini-2.5-pro`,
+    `gemini-3-pro-preview`, `gemini-3.1-pro-preview`, `deepseek-reasoner`,
+    `gpt-5.3-codex`, `gpt-5.2-codex`
 
 - **task_mode** (optional): Controls the system prompt persona. The calling LLM
   should choose based on the task:
@@ -596,7 +596,8 @@ models complex questions.
   million tokens for prompts ≤200k tokens, $4/$18 for prompts >200k tokens)
 - **deepseek-reasoner**: DeepSeek's reasoning model ($0.55/$2.19 per million
   tokens)
-- **gpt-5.2**: OpenAI's latest GPT model
+- **gpt-5.4**: OpenAI's GPT-5.4 model ($2.50/$15 per million tokens)
+- **gpt-5.2**: OpenAI's GPT-5.2 model ($1.75/$14 per million tokens)
 - **gpt-5.3-codex**: OpenAI's Codex model based on GPT-5.3
 - **gpt-5.2-codex**: OpenAI's Codex model based on GPT-5.2
 
@@ -703,12 +704,34 @@ for the full content.
 Save it as `~/.claude/commands/consult.md` and you can then use it by typing
 `/consult ask gemini about X` or `/consult ask codex about X` in Claude Code.
 
-## Debate skills
+## Multi-LLM skills
 
-Two skills that orchestrate structured debates between LLMs to find the best
-implementation approach before writing code. Both use `thread_id` to maintain
-conversation context across rounds, so each LLM remembers the full debate
-history without resending everything.
+Skills that orchestrate multi-turn conversations between LLMs. All use
+`thread_id` to maintain conversation context across rounds, so each LLM
+remembers the full history without resending everything.
+
+### collab
+
+**Collaborative ideation.** Gemini and Codex independently brainstorm ideas,
+then build on each other's suggestions across multiple rounds. Unlike debate,
+the tone is cooperative — refining and combining rather than critiquing. Claude
+synthesizes the strongest ideas into a plan and implements. See
+[skills/collab/SKILL.md](skills/collab/SKILL.md).
+
+```
+> /collab how should we handle offline sync for the mobile app
+```
+
+### collab-vs
+
+**Claude brainstorms with one LLM.** Claude and an opponent (Gemini or Codex)
+take turns building on each other's ideas. Like collab, but Claude participates
+directly instead of moderating. See
+[skills/collab-vs/SKILL.md](skills/collab-vs/SKILL.md).
+
+```
+> /collab-vs --gemini how should we handle offline sync for the mobile app
+```
 
 ### debate
 

package/dist/llm-cost.js

@@ -1,4 +1,8 @@
 const MODEL_PRICING = {
+    'gpt-5.4': {
+        inputCostPerMillion: 2.5,
+        outputCostPerMillion: 15.0,
+    },
     'gpt-5.2': {
         inputCostPerMillion: 1.75,
         outputCostPerMillion: 14.0,

package/dist/models.d.ts

@@ -1 +1 @@
-export declare const ALL_MODELS: readonly ["gemini-2.5-pro", "gemini-3-pro-preview", "gemini-3.1-pro-preview", "deepseek-reasoner", "gpt-5.2", "gpt-5.3-codex", "gpt-5.2-codex"];
+export declare const ALL_MODELS: readonly ["gemini-2.5-pro", "gemini-3-pro-preview", "gemini-3.1-pro-preview", "deepseek-reasoner", "gpt-5.2", "gpt-5.4", "gpt-5.3-codex", "gpt-5.2-codex"];

package/dist/models.js

@@ -4,6 +4,7 @@ export const ALL_MODELS = [
     'gemini-3.1-pro-preview',
     'deepseek-reasoner',
     'gpt-5.2',
+    'gpt-5.4',
     'gpt-5.3-codex',
     'gpt-5.2-codex',
 ];

package/dist/version.d.ts

@@ -1 +1 @@
-export declare const GIT_HASH = "1c9e0da";
+export declare const GIT_HASH = "c310fe4";

package/dist/version.js

@@ -1 +1 @@
-export const GIT_HASH = "1c9e0da";
+export const GIT_HASH = "c310fe4";

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "consult-llm-mcp",
-  "version": "2.4.2",
+  "version": "2.5.0",
   "description": "MCP server for consulting powerful AI models",
   "type": "module",
   "main": "dist/main.js",