claude-recall 0.11.3 → 0.11.4

package/.claude/skills/memory-management/SKILL.md

@@ -26,9 +26,10 @@ Persistent memory system that ensures Claude never repeats mistakes and always a
 
 1. **ALWAYS load rules before acting** - Call `load_rules` before Write, Edit, or significant Bash operations
 2. **Apply what you find** - Use retrieved preferences, patterns, and corrections
-3. **Capture corrections immediately** - User fixes are highest priority
-4. **Store learning cycles** - When you fail then succeed, that's valuable knowledge
-5. **Never store secrets** - No API keys, passwords, tokens, or PII
+3. **Ask before storing** - Before calling `store_memory`, briefly tell the user what you plan to store and ask for confirmation. Example: *"I'd like to remember that you prefer tabs over spaces. Store this?"* Only call the tool after the user agrees.
+4. **Capture corrections immediately** - User fixes are highest priority (still ask first)
+5. **Store learning cycles** - When you fail then succeed, that's valuable knowledge
+6. **Never store secrets** - No API keys, passwords, tokens, or PII
 
 ## Quick Reference
 
@@ -141,13 +142,17 @@ Safe to store:
 
 3. Fix the code
 
-4. Store the correction:
+4. Ask: "I'd like to remember: always use httpOnly cookies for auth tokens, never localStorage. Store this?"
+
+5. User: "Yes"
+
+6. Store the correction:
    mcp__claude-recall__store_memory({
      "content": "CORRECTION: Always use httpOnly cookies for auth tokens, never localStorage",
      "metadata": { "type": "correction" }
    })
 
-5. Response includes activeRule - apply it immediately
+7. Response includes activeRule - apply it immediately
 ```
 
 ### Overcoming a Challenge
@@ -160,7 +165,11 @@ Safe to store:
 
 3. Implemented JWT -> Works!
 
-4. Store the learning:
+4. Ask: "I'd like to remember: Redis sessions fail in k8s due to sync issues; use stateless JWT instead. Store this?"
+
+5. User: "Yes"
+
+6. Store the learning:
    mcp__claude-recall__store_memory({
      "content": "Auth in k8s: Redis sessions failed (sync issues). JWT stateless tokens work correctly.",
      "metadata": { "type": "failure", "learning_cycle": true }

package/README.md

@@ -17,6 +17,7 @@ Your preferences, project structure, workflows, corrections, and coding style ar
 
 - **Continuous Learning** — captures coding patterns, tool preferences, corrections, architectural decisions, and workflow habits in local SQLite
 - **Native Claude Skills** — no hooks required; Claude decides when to search/store based on built-in skill guidance
+- **User-Confirmed Storage** — Claude asks for your permission before storing any memory
 - **Project-Scoped Knowledge** — each project gets its own memory namespace; switch projects and Claude switches memory
 - **Failure Learning** — captures failures with counterfactual reasoning (what failed, why, what to do instead)
 - **Memory Evolution** — tracks agent progression over time across sophistication levels (L1–L4)
@@ -99,6 +100,7 @@ Exposes two tools to Claude Code:
 Installed automatically to `.claude/skills/memory-management/SKILL.md`. Teaches Claude:
 - Load rules before writing/editing code
 - Apply learned conventions and avoid past mistakes
+- **Ask the user for confirmation before storing any memory**
 - Capture corrections when users fix mistakes
 - Store learning cycles (fail → fix → success)
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-recall",
-  "version": "0.11.3",
+  "version": "0.11.4",
   "description": "Persistent memory for Claude Code with native Skills integration, automatic capture, failure learning, and project scoping via MCP server",
   "main": "dist/index.js",
   "bin": {