cc-reviewer 2.0.0 → 3.0.0

package/README.md

@@ -11,11 +11,11 @@ claude mcp add -s user cc-reviewer -- npx -y cc-reviewer
 
 **Step 2: Restart Claude Code**
 
-The MCP tools and slash commands (`/codex`, `/gemini`, `/multi`, `/ask-codex`, `/ask-gemini`, `/ask-multi`) are automatically installed.
+The MCP tools and slash commands (`/codex-review`, `/gemini-review`, `/multi-review`) are automatically installed.
 
 **Manual command install** (if needed):
 ```bash
-npx cc-reviewer --setup
+npx cc-reviewer update
 ```
 
 Verify with:
@@ -52,14 +52,10 @@ gemini  # follow auth prompts
 These tools provide **external second-opinion reviews** from Codex and Gemini CLIs. They are designed to complement Claude Code's native review capabilities, not replace them.
 
 **Review tools** (external second-opinion on your work):
-- `/codex` or "review with codex" - Get external Codex review
-- `/gemini` or "review with gemini" - Get external Gemini review
-- `/multi` - Get parallel reviews from both CLIs
-
-**Ask tools** (get help from a peer engineer):
-- `/ask-codex` - Ask Codex for help (planning, debugging, explaining, fixing)
-- `/ask-gemini` - Ask Gemini for help (architecture, patterns, scalability)
-- `/ask-multi` - Ask both models in parallel
+- `/codex-review` or "review with codex" - Get external Codex review
+- `/codex-xhigh-review` - Deep-thinking Codex review with xhigh reasoning
+- `/gemini-review` or "review with gemini" - Get external Gemini review
+- `/multi-review` - Get parallel reviews from both CLIs
 
 **For regular reviews:** Just say "review" and Claude Code will use its native capabilities. These external tools are only invoked when explicitly requested.
 
@@ -68,24 +64,18 @@ These tools provide **external second-opinion reviews** from Codex and Gemini CL
 These commands are available after restart:
 
 ```bash
-# Review tools
-/codex                    # Review with Codex
-/codex security           # Focus on security
-/codex-xhigh              # Codex with xhigh reasoning effort
-/gemini                   # Review with Gemini
-/gemini architecture      # Focus on architecture
-/multi                    # Both models in parallel
-
-# Ask tools (peer engineer)
-/ask-codex                # Ask Codex for help
-/ask-gemini               # Ask Gemini for help
-/ask-multi                # Ask both in parallel
+/codex-review             # Review with Codex
+/codex-review security    # Focus on security
+/codex-xhigh-review       # Codex with xhigh reasoning effort
+/gemini-review            # Review with Gemini
+/gemini-review architecture # Focus on architecture
+/multi-review             # Both models in parallel
 ```
 
 ## How It Works
 
 ```
-CC does work → User: /codex → External CLI reviews → CC synthesizes → Updated output
+CC does work → User: /codex-review → External CLI reviews → CC synthesizes → Updated output
 ```
 
 **Key Principles:**
@@ -108,33 +98,23 @@ CC does work → User: /codex → External CLI reviews → CC synthesizes → Up
 
 ## MCP Tools
 
-The plugin exposes six MCP tools:
+The plugin exposes three MCP tools:
 
 | Tool | Description |
 |------|-------------|
 | `codex_review` | Get Codex review (correctness, edge cases, performance) |
 | `gemini_review` | Get Gemini review (design patterns, scalability, tech debt) |
 | `multi_review` | Parallel review from both models |
-| `ask_codex` | Ask Codex for help (planning, debugging, explaining, fixing) |
-| `ask_gemini` | Ask Gemini for help (architecture, patterns, scalability) |
-| `ask_multi` | Ask both models in parallel |
 
 ## Output Format
 
-**Review tools** return structured JSON feedback with:
+**Review tools** return structured feedback from the external CLIs. Claude Code parses this feedback to identify:
 - **Findings**: Issues with severity, confidence, location, and suggestions
 - **Agreements**: Validations of CC's correct assessments
 - **Disagreements**: Challenges to CC's claims with corrections
 - **Alternatives**: Different approaches with tradeoffs
 - **Risk Assessment**: Overall risk level with top concerns
 
-**Ask tools** return structured JSON responses with:
-- **Answer**: Main response text (markdown)
-- **Key Points**: Bullet summary of main points
-- **Suggested Actions**: Recommended actions with priority and rationale
-- **File References**: Files examined with line ranges and relevance
-- **Alternatives**: Alternative approaches considered
-
 ## Development
 
 ```bash

package/commands/{codex-xhigh.md → codex-xhigh-review.md}

@@ -17,10 +17,10 @@ Use the `codex_review` MCP tool with `reasoningEffort: "xhigh"` for deeper analy
 2. Call the `codex_review` tool with:
    - `workingDir`: current working directory
    - `ccOutput`: brief summary of recent changes or context
-   - `reasoningEffort`: "xhigh" (this is the key difference from /codex)
+   - `reasoningEffort`: "xhigh" (this is the key difference from /codex-review)
    - `serviceTier`: if user says "fast mode"/"fast"/"priority" → "fast"; if "flex"/"cheap"/"budget" → "flex"; otherwise omit
-   - `focus`: extracted from $ARGUMENTS if it's a known focus area
-   - `customInstructions`: $ARGUMENTS if it's custom text
+   - `focusAreas`: extracted from $ARGUMENTS if it's a known focus area
+   - `customPrompt`: $ARGUMENTS if it's custom text
 
 3. After receiving feedback - VALIDATE before accepting:
 

package/commands/{multi.md → multi-review.md}

@@ -7,7 +7,7 @@ Get parallel reviews from both Codex and Gemini, raw output for manual synthesis
 
 ## When to Use
 
-Use `/multi` when you want parallel reviews from both Codex and Gemini.
+Use `/multi-review` when you want parallel reviews from both Codex and Gemini.
 
 ## Before Calling - PREPARE THE HANDOFF
 

package/dist/adapters/base.d.ts

@@ -5,8 +5,7 @@
  * Makes it easy to add new models (Ollama, Azure, etc.) without
  * changing the core orchestration logic.
  */
-import { ReviewOutput, PeerOutput } from '../schema.js';
-import { FocusArea, OutputType, ReasoningEffort, ServiceTier, TaskType } from '../types.js';
+import { FocusArea, OutputType, ReasoningEffort, ServiceTier } from '../types.js';
 export interface ReviewerCapabilities {
     /** Display name for this reviewer */
     name: string;
@@ -55,30 +54,9 @@ export interface ExpertRole {
 export declare const EXPERT_ROLES: Record<string, ExpertRole>;
 /** @deprecated Use handoff.ts selectRole() instead */
 export declare function selectExpertRole(focusAreas?: FocusArea[]): ExpertRole;
-export interface PeerRequest {
-    /** Working directory containing the code */
-    workingDir: string;
-    /** The question or request from CC */
-    prompt: string;
-    /** Hint about the type of task */
-    taskType?: TaskType;
-    /** Files the peer should focus on */
-    relevantFiles?: string[];
-    /** Additional context (error messages, prior analysis) */
-    context?: string;
-    /** Areas to focus on */
-    focusAreas?: FocusArea[];
-    /** Custom instructions from the user */
-    customPrompt?: string;
-    /** Reasoning effort level (for models that support it) */
-    reasoningEffort?: ReasoningEffort;
-    /** Service tier (for models that support it: priority = fast, flex = cheap) */
-    serviceTier?: ServiceTier;
-}
 export interface ReviewSuccess {
     success: true;
-    output: ReviewOutput;
-    rawOutput?: string;
+    output: string;
     executionTimeMs: number;
 }
 export interface ReviewFailure {
@@ -94,20 +72,6 @@ export interface ReviewError {
     message: string;
     details?: Record<string, unknown>;
 }
-export interface PeerSuccess {
-    success: true;
-    output: PeerOutput;
-    rawOutput?: string;
-    executionTimeMs: number;
-}
-export interface PeerFailure {
-    success: false;
-    error: ReviewError;
-    suggestion?: string;
-    rawOutput?: string;
-    executionTimeMs: number;
-}
-export type PeerResult = PeerSuccess | PeerFailure;
 /**
  * Base interface that all reviewer adapters must implement.
  * This allows easy addition of new AI CLIs without changing orchestration logic.
@@ -121,13 +85,11 @@ export interface ReviewerAdapter {
     isAvailable(): Promise<boolean>;
     /** Run a review and return structured output */
     runReview(request: ReviewRequest): Promise<ReviewResult>;
-    /** Run a general-purpose peer request and return structured output */
-    runPeerRequest(request: PeerRequest): Promise<PeerResult>;
     /**
      * Optional: Run peer review of another model's output
      * Future capability - not currently implemented by any adapter
      */
-    runPeerReview?(originalRequest: ReviewRequest, reviewToScore: ReviewOutput): Promise<ReviewResult>;
+    runPeerReview?(originalRequest: ReviewRequest, reviewToScore: string): Promise<ReviewResult>;
 }
 export declare function registerAdapter(adapter: ReviewerAdapter): void;
 export declare function getAdapter(id: string): ReviewerAdapter | undefined;

package/dist/adapters/codex.d.ts

@@ -2,20 +2,18 @@
  * Codex CLI Adapter
  *
  * Implements the ReviewerAdapter interface for OpenAI's Codex CLI.
- * Specializes in correctness, edge cases, and performance analysis.
+ * Returns raw text — no JSON parsing or schema enforcement.
+ * CC handles interpretation of the reviewer's response.
  */
-import { ReviewerAdapter, ReviewerCapabilities, ReviewRequest, ReviewResult, PeerRequest, PeerResult } from './base.js';
+import { ReviewerAdapter, ReviewerCapabilities, ReviewRequest, ReviewResult } from './base.js';
 export declare class CodexAdapter implements ReviewerAdapter {
     readonly id = "codex";
     getCapabilities(): ReviewerCapabilities;
     isAvailable(): Promise<boolean>;
     runReview(request: ReviewRequest): Promise<ReviewResult>;
-    private runWithRetry;
-    runPeerRequest(request: PeerRequest): Promise<PeerResult>;
-    private runPeerWithRetry;
     private runCli;
+    private handleException;
     private categorizeError;
     private getSuggestion;
-    private parseRetryAfter;
 }
 export declare const codexAdapter: CodexAdapter;