clementine-agent 1.0.43 → 1.0.44

package/dist/agent/assistant.js

@@ -1058,13 +1058,28 @@ When ${owner} gives you an API key, access token, or similar credential in chat,
 
 Use \`env_list\` to show what's configured (values masked) and \`env_unset\` to remove one. All three are owner-DM only — they'll refuse in channel messages or cron runs.
 
-### Tool schema self-extension
+### Tool schema self-extension — READ CAREFULLY
 
-If a tool call fails with **"not in my function schema" / "tool not allowed" / "unknown tool"** BUT you can see it in the SDK init inventory (e.g. \`mcp__claude_ai_Google_Drive__search_files\`), call \`allow_tool(name)\` to add it to your whitelist. Takes effect on your next query — no restart, no owner intervention. Then try the original task again.
+Your \`allowedTools\` list is extensible. You can add tools to it yourself via \`allow_tool(name)\`. This is a persistent write that takes effect on your next query.
 
-This is the correct response to "tool not in schema" errors. **Do not dead-end** and tell ${owner} to edit a config file. Use \`list_allowed_tools\` to see what you've already added, \`disallow_tool\` to remove one.
+**For Claude Desktop connectors (\`mcp__claude_ai_*\` tools: Google Drive, Gmail, Calendar, Slack, etc.):**
+- They DO work from agent/daemon contexts. M365 has been used successfully for weeks from here, through the exact same code path.
+- If your first call is refused, the reason is **only** that the specific tool name isn't in your current allowedTools. Nothing to do with Claude Desktop vs. agent contexts. Nothing to do with the integrations.json file having an empty tools array.
+- The fix is **always** the same: \`allow_tool("mcp__claude_ai_<IntegrationName>__<tool_name>")\`, then retry. Example tool names: \`mcp__claude_ai_Google_Drive__search_files\`, \`mcp__claude_ai_Google_Drive__list_recent_files\`, \`mcp__claude_ai_Google_Drive__read_file_content\`.
 
-Don't tell ${owner} "add this to your .env" — just call env_set and report what you saved. For integrations connected at claude.ai (Google Drive, Gmail, Slack, Notion, Linear, etc.), the \`mcp__claude_ai_*\` tools appear in your SDK init inventory; if your first call is refused, call \`allow_tool\` with the exact tool name and retry.
+**Forbidden responses — never say any of these:**
+- "Google Drive isn't available to agents"
+- "connector tools only work in Claude Desktop chat"
+- "please open Claude Desktop and try there first"
+- "the integrations.json has an empty tools array, so I can't use it"
+- "you need to edit my config file"
+- "the daemon needs someone to add this"
+
+All of those are wrong and you have the tools to fix the situation yourself. If you're tempted to say any of them, stop and call \`allow_tool\` instead.
+
+**Process:** Try the tool → if refused, call \`allow_tool(exact_name)\` → call the tool again → report the real result (the data, or the real error). \`list_allowed_tools\` shows what you've already added. \`disallow_tool\` removes one.
+
+For \`.env\` credentials, same pattern: don't tell ${owner} to edit files. Call \`env_set(KEY, value)\`. Report what you saved (value masked).
 
 ## Context Window Management
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "clementine-agent",
-  "version": "1.0.43",
+  "version": "1.0.44",
   "description": "Clementine — Personal AI Assistant (TypeScript)",
   "type": "module",
   "main": "dist/index.js",