rrce-workflow 0.3.6 → 0.3.7

package/dist/index.js

@@ -1107,6 +1107,33 @@ import { stringify } from "yaml";
 function getOpenCodeConfigPath() {
   return path8.join(os2.homedir(), ".config", "opencode", "opencode.json");
 }
+function enableProviderCaching() {
+  const opencodePath = getOpenCodeConfigPath();
+  let config = {};
+  if (fs7.existsSync(opencodePath)) {
+    try {
+      config = JSON.parse(fs7.readFileSync(opencodePath, "utf8"));
+    } catch (e) {
+      console.error("Warning: Could not parse existing OpenCode config, creating new provider section");
+    }
+  } else {
+    ensureDir(path8.dirname(opencodePath));
+  }
+  if (!config.provider) {
+    config.provider = {};
+  }
+  const providers = ["anthropic", "openai", "openrouter", "google"];
+  for (const provider of providers) {
+    if (!config.provider[provider]) {
+      config.provider[provider] = {};
+    }
+    if (!config.provider[provider].options) {
+      config.provider[provider].options = {};
+    }
+    config.provider[provider].options.setCacheKey = true;
+  }
+  fs7.writeFileSync(opencodePath, JSON.stringify(config, null, 2));
+}
 function copyPromptsToDir(prompts, targetDir, extension) {
   for (const prompt of prompts) {
     const baseName = path8.basename(prompt.filePath, ".md");
@@ -6169,6 +6196,7 @@ async function installAgentPrompts(config, workspacePath, dataPaths) {
       const primaryDataPath = dataPaths[0];
       if (primaryDataPath) {
         surgicalUpdateOpenCodeAgents(prompts, config.storageMode, primaryDataPath);
+        enableProviderCaching();
       }
     }
   }

package/docs/MIGRATION-v2.md

@@ -94,31 +94,36 @@ Orchestrator: (Detects user wants implementation, auto-proceeds to planning →
 
 ---
 
-### 4. Model Configuration (Cost Optimization)
+### 4. Provider Caching Configuration (Model-Agnostic)
 
-**What changed:** Research agent now uses **Claude Haiku** by default (12x cheaper)
+**What changed:** RRCE now enables provider caching for ALL providers, but does NOT force specific models
 
-| Agent | Old Model | New Model | Cost Change |
-|-------|-----------|-----------|-------------|
-| Research | Sonnet 4 | Haiku 4 | -92% |
-| Planning | Sonnet 4 | Sonnet 4 | No change |
-| Executor | Sonnet 4 | Sonnet 4 | No change |
+**New behavior:**
+- `setCacheKey: true` enabled for Anthropic, OpenAI, OpenRouter, Google
+- **No hardcoded models** - you choose what model to use
+- Prompt caching activates automatically for any model that supports it
 
-**Impact:** Research phase is now 98% cheaper, with minimal quality difference for Q&A tasks
+**Recommended models** (optional, for cost optimization):
 
-**Migration action:** None required (configured in `opencode.json`)
+| Agent | Recommendation | Rationale |
+|-------|----------------|-----------|
+| Research | Haiku/Mini | Q&A doesn't need heavy reasoning |
+| Planning | Sonnet/GPT-4o | Task breakdown needs reasoning |
+| Executor | Sonnet/GPT-4o | Code generation needs power |
 
-**To revert to Sonnet for research:**
+**To set models per agent (optional):**
 ```json
 {
   "agent": {
     "rrce_research_discussion": {
-      "model": "anthropic/claude-sonnet-4-20250514"
+      "model": "anthropic/claude-haiku-4-20250514"
     }
   }
 }
 ```
 
+**Migration action:** None required - caching is automatic
+
 ---
 
 ### 5. Session State Management

package/docs/opencode-guide-optimization-addendum.md

@@ -138,22 +138,39 @@ Turn 3: Reference cached findings → Generate brief
 
 ---
 
-## 💰 Model Configuration
-
-Optimized model selection for balanced cost/quality:
-
-| Agent | Model | Use Case | Cost |
-|-------|-------|----------|------|
-| **Research** | Claude Haiku 4 | Q&A, clarification | $0.25/M tokens |
-| **Planning** | Claude Sonnet 4 | Task breakdown, reasoning | $3/M tokens |
-| **Executor** | Claude Sonnet 4 | Code generation | $3/M tokens |
-| **Orchestrator** | Claude Sonnet 4 | Multi-phase coordination | $3/M tokens |
-
-**Why Haiku for Research?**
-- Research is primarily Q&A and documentation
-- Haiku is 12x cheaper than Sonnet
-- Quality is sufficient for clarification questions
-- Complex reasoning happens in Planning/Execution (Sonnet)
+## 💰 Model Configuration (User Choice)
+
+**RRCE does NOT force specific models.** You choose what works for you!
+
+The optimization works with **ANY model** you configure in OpenCode. What we enable:
+
+| Feature | What It Does | Works With |
+|---------|--------------|------------|
+| **Provider Caching** | `setCacheKey: true` for all providers | Anthropic, OpenAI, OpenRouter, Google |
+| **Prompt Caching** | Automatic via `promptCacheKey = sessionID` | Any model that supports caching |
+| **Slim Prompts** | 70-93% smaller prompts | ALL models |
+| **Session Reuse** | Orchestrator reuses sessions | ALL models |
+
+### Recommended Models (Optional)
+
+If you want cost optimization, consider:
+
+| Agent | Recommendation | Rationale |
+|-------|----------------|-----------|
+| **Research** | Haiku/Mini | Q&A doesn't need heavy reasoning |
+| **Planning** | Sonnet/GPT-4o | Task breakdown needs reasoning |
+| **Executor** | Sonnet/GPT-4o | Code generation needs power |
+
+**To set models per agent (optional):**
+```json
+{
+  "agent": {
+    "rrce_research_discussion": {
+      "model": "anthropic/claude-haiku-4-20250514"
+    }
+  }
+}
+```
 
 **Configuration location:** `opencode.json` in project root or `~/.config/opencode/opencode.json`
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "rrce-workflow",
-  "version": "0.3.6",
+  "version": "0.3.7",
   "description": "RRCE-Workflow TUI - Agentic code workflow generator for AI-assisted development",
   "author": "RRCE Team",
   "license": "MIT",