chore: version up to 2.80.34 and package

package.json

@@ -2,7 +2,7 @@
   "name": "astra",
   "displayName": "Astra",
   "description": "The personal intelligence layer for Antigravity and VS Code. A private cognitive partner for deep project context, memory, and proactive strategic decision-making.",
-  "version": "2.80.33",
+  "version": "2.80.34",
   "publisher": "g1nation",
   "license": "MIT",
   "icon": "assets/icon.png",
@@ -178,6 +178,45 @@
           "default": 300,
           "description": "Request timeout in seconds. Default: 300"
         },
+        "g1nation.contextLength": {
+          "type": "number",
+          "default": 32768,
+          "minimum": 2048,
+          "description": "Model context window in tokens (prompt + generation combined). Set this to the value your loaded model is actually running with in LM Studio / Ollama. Astra budgets prompt and output against this so it never overflows. Default: 32768"
+        },
+        "g1nation.maxOutputTokens": {
+          "type": "number",
+          "default": 4096,
+          "minimum": 256,
+          "description": "Upper bound on tokens generated per response. The effective limit is reduced automatically when the prompt is large so input + output stays within g1nation.contextLength. Default: 4096"
+        },
+        "g1nation.contextSafetyMargin": {
+          "type": "number",
+          "default": 2048,
+          "minimum": 0,
+          "description": "Tokens kept free as a safety buffer for token-count estimation error. Default: 2048"
+        },
+        "g1nation.contextOverflowPolicy": {
+          "type": "string",
+          "enum": [
+            "stopAtLimit",
+            "truncateMiddle",
+            "rollingWindow"
+          ],
+          "default": "stopAtLimit",
+          "description": "Fallback behavior (LM Studio) if the prompt still exceeds the context window after Astra's own budgeting. 'stopAtLimit' fails clearly so you notice; 'truncateMiddle'/'rollingWindow' drop content silently. Default: stopAtLimit"
+        },
+        "g1nation.autoCompactHistory": {
+          "type": "boolean",
+          "default": true,
+          "description": "Automatically drop the oldest conversation messages from the request when the prompt would exceed the context budget (the on-screen chat history is unaffected). Default: true"
+        },
+        "g1nation.smallModelContextCap": {
+          "type": "number",
+          "default": 8192,
+          "minimum": 0,
+          "description": "When a small model (≤4B parameters, detected from the model name) is selected, budget the prompt against this smaller effective context window instead of g1nation.contextLength — small models often emit an empty/EOS response on prompts that nominally fit but exceed their real capability. Set 0 to disable. Default: 8192"
+        },
         "g1nation.lmStudio.idleTimeoutMs": {
           "type": "number",
           "default": 300000,
@@ -292,6 +331,11 @@
           "type": "object",
           "default": {},
           "description": "Inline fallback for the agent ↔ knowledge mapping. Used only when the JSON file is missing. Shape: { defaultAgent?, agents: [{ name, knowledgeFolders, model?, description? }] }. Folder paths can be absolute, ~-prefixed, or relative to the active brain root."
+        },
+        "g1nation.agentSkillsPath": {
+          "type": "string",
+          "default": "",
+          "description": "Absolute path to the agent skills folder (`.agent/skills/*.md`). When empty, defaults to '<workspace>/.agent/skills'. Use this on Windows or when your skills live outside the workspace."
         }
       }
     }