fixes

ai_client.py

@@ -1,4 +1,4 @@
-# ai_client.py
+# ai_client.py
 import tomllib
 import json
 import datetime
@@ -318,7 +318,7 @@ def _send_gemini(md_content: str, user_message: str, base_dir: str) -> str:
         full_message = f"<context>\n{md_content}\n</context>\n\n{user_message}"
 
         _append_comms("OUT", "request", {
-            "message": full_message,
+            "summary": f"[Gemini: context {len(md_content)} chars + user message {len(user_message)} chars]",
         })
 
         response = _gemini_chat.send_message(full_message)
@@ -546,6 +546,21 @@ def _build_files_context_blocks(
     return blocks
 
 
+
+def _strip_cache_controls(history: list[dict]):
+    """
+    Remove cache_control from all content blocks in the message history.
+    Anthropic allows a maximum of 4 cache_control blocks total across
+    system + tools + messages.  We reserve those slots for the stable
+    system/tools prefix and the current turn's context block, so all
+    older history entries must be clean.
+    """
+    for msg in history:
+        content = msg.get("content")
+        if isinstance(content, list):
+            for block in content:
+                if isinstance(block, dict):
+                    block.pop("cache_control", None)
 def _run_anthropic_loop(
     user_content: list[dict],
     user_message: str,
@@ -559,6 +574,7 @@ def _run_anthropic_loop(
     """
     global _anthropic_history
 
+    _strip_cache_controls(_anthropic_history)
     _anthropic_history.append({"role": "user", "content": user_content})
 
     _append_comms("OUT", "request", {
@@ -779,4 +795,5 @@ def send(
         return _send_gemini(md_content, user_message, base_dir)
     elif _provider == "anthropic":
         return _send_anthropic(md_content, user_message, base_dir, file_items)
-    raise ValueError(f"unknown provider: {_provider}")
+    raise ValueError(f"unknown provider: {_provider}")
+

config.toml

@@ -23,12 +23,10 @@ paths = []
 
 [discussion]
 history = [
-    "[User]\nAdd support for properly handling files that we have in config with anthropics api (your context will have documentation reguarding it). This will allow us to free up the main markdown file were sending as a message.\nWe also want to make sure the messaging of that giant markdown is chunked as anthropic does have an upper limit per messages.",
+    "User:\nAdd support for properly handling files that we have in config with anthropics api (your context will have documentation reguarding it). This will allow us to free up the main markdown file were sending as a message.\nWe also want to make sure the messaging of that giant markdown is chunked as anthropic does have an upper limit per messages.\nAI:\nGood. Now the big rewrite of `ai_client.py`. The strategy:\n\n- Add a new `_send_anthropic_files` path that uses `file_cache.get_file_id()` to upload files individually as `document`/`image` content blocks, keeping the context message lean\n- The fallback (and also for oversized content) uses chunked text blocks — split the markdown into ≤180k-char chunks (well under the ~200k token limit with headroom for overhead), each sent as a separate `text` content block with `cache_control` on the last one\n- `send()` now accepts an optional `file_items` parameter from aggregate\n- `reset_session()` also resets the file_cache client\nUser:\ncontinue. Also got this error when trying to run latest version of this program. However I can use an old version for now.\nAi:\nNow fix `gui.py` — patch the three 2-tuple unpack sites and thread `file_items` through to `send()`:\n[User]\ncontinue. My bad the error in comms was shown int he screenshot I linked. in the md context. Make sure to update MainContext.md when done.\nAI:\nfinished without sending a response...\nUser:\nSeems to be working...! \nMake a richer discussioon history editor, with structual support for the user/ai message nodes and buttons in the reponse and message panels for adding their entries to it on click.\nAI:\nThe existing format already uses `[User]\\n...` and `[AI]\\n...` prefixes. The new structured editor will parse those naturally and write them back in the same format. Now let me implement the full changes:\nUser:\nYou had failres in your last few called scripts. Git diff shows now changes to python files.\nI'll add the comms and tool logs to your context. \nWe should remove the script content from the tool log and just save the script name so that it doesn't bloat it. \nSame with omitting full message sent in the comms log, very expensive to send to you or read manually.\nVendor API:\nERROR: Error code: 400 - {'type': 'error', 'error': {'type': 'invalid_request_error', 'message': 'A maximum of 4 blocks with cache_control may be provided. Found 5.'}, 'request_id': 'req_011CYMxgMXvcZUVFNyFz1Lx8'}",
-    "[AI]\nGood. Now the big rewrite of `ai_client.py`. The strategy:\n\n- Add a new `_send_anthropic_files` path that uses `file_cache.get_file_id()` to upload files individually as `document`/`image` content blocks, keeping the context message lean\n- The fallback (and also for oversized content) uses chunked text blocks — split the markdown into ≤180k-char chunks (well under the ~200k token limit with headroom for overhead), each sent as a separate `text` content block with `cache_control` on the last one\n- `send()` now accepts an optional `file_items` parameter from aggregate\n- `reset_session()` also resets the file_cache client",
+    "[User]:\nI'm not able to send a message for a while now...\n(Message finally sent)",
-    "[User]\ncontinue. Also got this error when trying to run latest version of this program. However I can use an old version for now.",
+    "[AI]\nCompleted without a response message showing up.",
-    "[Ai]\nNow fix `gui.py` — patch the three 2-tuple unpack sites and thread `file_items` through to `send()`:",
+    "[User]\nRestarts custom prompting program.",
-    "[User]\ncontinue. My bad the error in comms was shown int he screenshot I linked. in the md context. Make sure to update MainContext.md when done.",
-    "[AI]\nfinished without sending a repose...",
 ]
 
 [ai]

dpg_layout.ini

@@ -107,7 +107,7 @@ DockId=0x0000001F,0
 
 [Window][###106]
 Pos=551,0
-Size=1544,2137
+Size=1275,2137
 Collapsed=0
 DockId=0x00000012,2
 
@@ -174,7 +174,7 @@ DockId=0x00000016,0
 
 [Window][###112]
 Pos=551,0
-Size=1544,2137
+Size=1275,2137
 Collapsed=0
 DockId=0x00000012,1
 
@@ -321,26 +321,26 @@ Collapsed=0
 DockId=0x00000026,0
 
 [Window][###82]
-Pos=2097,0
+Pos=1828,0
-Size=1743,1162
+Size=2012,1162
 Collapsed=0
 DockId=0x0000001F,0
 
 [Window][###89]
 Pos=551,0
-Size=1544,2137
+Size=1275,2137
 Collapsed=0
 DockId=0x00000012,0
 
 [Window][###97]
-Pos=2097,1649
+Pos=1828,1649
-Size=1743,488
+Size=2012,488
 Collapsed=0
 DockId=0x00000028,0
 
 [Window][###104]
-Pos=2097,1164
+Pos=1828,1164
-Size=1743,483
+Size=2012,483
 Collapsed=0
 DockId=0x00000027,0
 
@@ -404,6 +404,31 @@ Pos=1578,868
 Size=700,440
 Collapsed=0
 
+[Window][###139]
+Pos=1578,868
+Size=700,440
+Collapsed=0
+
+[Window][###241]
+Pos=1578,868
+Size=700,440
+Collapsed=0
+
+[Window][###346]
+Pos=1578,868
+Size=700,440
+Collapsed=0
+
+[Window][###457]
+Pos=1578,868
+Size=700,440
+Collapsed=0
+
+[Window][###577]
+Pos=1578,868
+Size=700,440
+Collapsed=0
+
 [Docking][Data]
 DockSpace                     ID=0x7C6B3D9B Window=0xA87D555D Pos=0,0 Size=3840,2137 Split=X Selected=0x40484D8F
   DockNode                    ID=0x00000003 Parent=0x7C6B3D9B SizeRef=549,1161 Split=Y Selected=0xEE087978

session_logger.py

@@ -123,3 +123,4 @@ def log_tool_call(script: str, result: str, script_path: str | None):
         pass
 
     return str(ps1_path) if ps1_path else None
+