ed/manual_slop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

amazing

Browse Source
This commit is contained in:
Edward R. Gonzalez
2026-02-21 15:35:44 -05:00
parent 308070e885
commit 105a80bbe7
5 changed files with 326 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
MainContext.md
View File

@@ -11,7 +11,7 @@
**Files:**
- `gui.py` - main GUI, `App` class, all panels, all callbacks, confirmation dialog, layout persistence
- `ai_client.py` - unified provider wrapper, model listing, session management, send, tool/function-call loop
- `ai_client.py` - unified provider wrapper, model listing, session management, send, tool/function-call loop, comms log
- `aggregate.py` - reads config, collects files/screenshots/discussion, writes numbered `.md` files to `output_dir`
- `shell_runner.py` - subprocess wrapper that runs PowerShell scripts sandboxed to `base_dir`, returns stdout/stderr/exit code as a string
- `config.toml` - namespace, output_dir, files paths+base_dir, screenshots paths+base_dir, discussion history array, ai provider+model
@@ -23,10 +23,11 @@
- **Files** - base_dir, scrollable path list with remove, add file(s), add wildcard
- **Screenshots** - base_dir, scrollable path list with remove, add screenshot(s)
- **Discussion History** - multiline text box, `---` as separator between excerpts, save splits on `---` back into toml array
- **Provider** - provider combo (gemini/anthropic), model listbox populated from API, fetch models button, status line
- **Provider** - provider combo (gemini/anthropic), model listbox populated from API, fetch models button
- **Message** - multiline input, Gen+Send button, MD Only button, Reset session button
- **Response** - readonly multiline displaying last AI response
- **Tool Calls** - scrollable log of every PowerShell tool call the AI made, showing script and result; Clear button
- **Comms History** - live log of every raw request/response/tool_call/tool_result exchanged with the vendor API; status line lives here; Clear button; heavy fields (message, text, script, output) clamped to an 80px scrollable box when they exceed `COMMS_CLAMP_CHARS` (300) characters
**Layout persistence:**
- `dpg.configure_app(..., init_file="dpg_layout.ini")` loads the ini at startup if it exists; DPG silently ignores a missing file
@@ -42,11 +43,26 @@
- Before any script runs, `gui.py` shows a modal `ConfirmDialog` on the main thread; the background send thread blocks on a `threading.Event` until the user clicks Approve or Reject
- The dialog displays `base_dir`, shows the script in an editable text box (allowing last-second tweaks), and has Approve & Run / Reject buttons
- On approval the (possibly edited) script is passed to `shell_runner.run_powershell()` which prepends `Set-Location -LiteralPath '<base_dir>'` and runs it via `powershell -NoProfile -NonInteractive -Command`
- Every script (original, before Set-Location is prepended) is saved to ./scripts/generated/ai_<timestamp>.ps1 before execution; the saved path appears in the tool result
- stdout, stderr, and exit code are returned to the AI as the tool result
- Rejections return `"USER REJECTED: command was not executed"` to the AI
- All tool calls (script + result/rejection) are appended to `_tool_log` and displayed in the Tool Calls panel
**Comms Log (ai_client.py):**
- `_comms_log: list[dict]` accumulates every API interaction during a session
- `_append_comms(direction, kind, payload)` called at each boundary: OUT/request before sending, IN/response after each model reply, OUT/tool_call before executing, IN/tool_result after executing, OUT/tool_result_send when returning results to the model
- Entry fields: `ts` (HH:MM:SS), `direction` (OUT/IN), `kind`, `provider`, `model`, `payload` (dict)
- Anthropic responses also include `usage` (input_tokens/output_tokens) and `stop_reason` in payload
- `get_comms_log()` returns a snapshot; `clear_comms_log()` empties it
- `comms_log_callback` (injected by gui.py) is called from the background thread with each new entry; gui queues entries in `_pending_comms` (lock-protected) and flushes them to the DPG panel each render frame
- `MAX_FIELD_CHARS = 400` in ai_client is the threshold used for the clamp decision in the UI (`COMMS_CLAMP_CHARS = 300` in gui.py governs the display cutoff)
**Comms History panel rendering:**
- Each entry shows: index, timestamp, direction (colour-coded blue=OUT / green=IN), kind (colour-coded), provider/model
- Payload fields rendered below the header; fields in `_HEAVY_KEYS` (`message`, `text`, `script`, `output`, `content`) that exceed `COMMS_CLAMP_CHARS` are shown in an 80px tall readonly scrollable `input_text` box instead of a plain `add_text`
- Colour legend row at the top of the panel
- Status line (formerly in Provider panel) moved to top of Comms History panel
- Reset session also clears the comms log and panel; Clear button in Comms History clears only the comms log
**Data flow:**
1. GUI edits are held in `App` state lists (`self.files`, `self.screenshots`, `self.history`) and dpg widget values
2. `_flush_to_config()` pulls all widget values into `self.config` dict
@@ -54,7 +70,7 @@
4. `cb_generate_send()` calls `_do_generate()` then threads a call to `ai_client.send(md, message, base_dir)`
5. `ai_client.send()` prepends the md as a `<context>` block to the user message and sends via the active provider chat session
6. If the AI responds with tool calls, the loop handles them (with GUI confirmation) before returning the final text response
7. Sessions are stateful within a run (chat history maintained), `Reset` clears them and the tool log
7. Sessions are stateful within a run (chat history maintained), `Reset` clears them, the tool log, and the comms log
**Config persistence:**
- Every send and save writes `config.toml` with current state including selected provider and model under `[ai]`
@@ -66,9 +82,11 @@
- AI sends and model fetches run on daemon background threads
- `_pending_dialog` (guarded by a `threading.Lock`) is set by the background thread and consumed by the render loop each frame, calling `dialog.show()` on the main thread
- `dialog.wait()` blocks the background thread on a `threading.Event` until the user acts
- `_pending_comms` (guarded by a separate `threading.Lock`) is populated by `_on_comms_entry` (background thread) and drained by `_flush_pending_comms()` each render frame (main thread)
**Known extension points:**
- Add more providers by adding a section to `credentials.toml`, a `_list_*` and `_send_*` function in `ai_client.py`, and the provider name to the `PROVIDERS` list in `gui.py`
- System prompt support could be added as a field in `config.toml` and passed in `ai_client.send()`
- Discussion history excerpts could be individually toggleable for inclusion in the generated md
- `MAX_TOOL_ROUNDS` in `ai_client.py` caps agentic loops at 5 rounds; adjustable
- `COMMS_CLAMP_CHARS` in `gui.py` controls the character threshold for clamping heavy payload fields

116
ai_client.py
View File

@@ -1,5 +1,7 @@
# ai_client.py
import tomllib
import json
import datetime
from pathlib import Path
_provider: str = "gemini"
@@ -16,8 +18,56 @@ _anthropic_history: list[dict] = []
# Returns the output string if approved, None if rejected.
confirm_and_run_callback = None
# Injected by gui.py - called whenever a comms entry is appended.
# Signature: (entry: dict) -> None
comms_log_callback = None
MAX_TOOL_ROUNDS = 5
# ------------------------------------------------------------------ comms log
_comms_log: list[dict] = []
MAX_FIELD_CHARS = 400 # beyond this we show a truncated preview in the UI
def _clamp(value, max_chars: int = MAX_FIELD_CHARS) -> tuple[str, bool]:
"""Return (display_str, was_truncated)."""
if isinstance(value, (dict, list)):
s = json.dumps(value, ensure_ascii=False, indent=2)
else:
s = str(value)
if len(s) > max_chars:
return s[:max_chars], True
return s, False
def _append_comms(direction: str, kind: str, payload: dict):
"""
direction : "OUT" | "IN"
kind : "request" | "response" | "tool_call" | "tool_result"
payload : raw dict describing the event
"""
entry = {
"ts": datetime.datetime.now().strftime("%H:%M:%S"),
"direction": direction,
"kind": kind,
"provider": _provider,
"model": _model,
"payload": payload,
}
_comms_log.append(entry)
if comms_log_callback is not None:
comms_log_callback(entry)
def get_comms_log() -> list[dict]:
return list(_comms_log)
def clear_comms_log():
_comms_log.clear()
def _load_credentials() -> dict:
with open("credentials.toml", "rb") as f:
return tomllib.load(f)
@@ -264,15 +314,33 @@ 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,
})
response = _gemini_chat.send_message(full_message)
for _ in range(MAX_TOOL_ROUNDS):
for round_idx in range(MAX_TOOL_ROUNDS):
# Log the raw response candidates as text summary
text_parts_raw = [
part.text
for candidate in response.candidates
for part in candidate.content.parts
if hasattr(part, "text") and part.text
]
tool_calls = [
part.function_call
for candidate in response.candidates
for part in candidate.content.parts
if part.function_call is not None
]
_append_comms("IN", "response", {
"round": round_idx,
"text": "\n".join(text_parts_raw),
"tool_calls": [{"name": fc.name, "args": dict(fc.args)} for fc in tool_calls],
})
if not tool_calls:
break
@@ -280,7 +348,15 @@ def _send_gemini(md_content: str, user_message: str, base_dir: str) -> str:
for fc in tool_calls:
if fc.name == TOOL_NAME:
script = fc.args.get("script", "")
_append_comms("OUT", "tool_call", {
"name": TOOL_NAME,
"script": script,
})
output = _run_script(script, base_dir)
_append_comms("IN", "tool_result", {
"name": TOOL_NAME,
"output": output,
})
function_responses.append(
types.Part.from_function_response(
name=TOOL_NAME,
@@ -325,7 +401,11 @@ def _send_anthropic(md_content: str, user_message: str, base_dir: str) -> str:
full_message = f"<context>\n{md_content}\n</context>\n\n{user_message}"
_anthropic_history.append({"role": "user", "content": full_message})
for _ in range(MAX_TOOL_ROUNDS):
_append_comms("OUT", "request", {
"message": full_message,
})
for round_idx in range(MAX_TOOL_ROUNDS):
response = _anthropic_client.messages.create(
model=_model,
max_tokens=8096,
@@ -338,6 +418,24 @@ def _send_anthropic(md_content: str, user_message: str, base_dir: str) -> str:
"content": response.content
})
# Summarise the response content for the log
text_blocks = [b.text for b in response.content if hasattr(b, "text") and b.text]
tool_use_blocks = [
{"id": b.id, "name": b.name, "input": b.input}
for b in response.content
if b.type == "tool_use"
]
_append_comms("IN", "response", {
"round": round_idx,
"stop_reason": response.stop_reason,
"text": "\n".join(text_blocks),
"tool_calls": tool_use_blocks,
"usage": {
"input_tokens": response.usage.input_tokens,
"output_tokens": response.usage.output_tokens,
} if response.usage else {},
})
if response.stop_reason != "tool_use":
break
@@ -345,7 +443,17 @@ def _send_anthropic(md_content: str, user_message: str, base_dir: str) -> str:
for block in response.content:
if block.type == "tool_use" and block.name == TOOL_NAME:
script = block.input.get("script", "")
_append_comms("OUT", "tool_call", {
"name": TOOL_NAME,
"id": block.id,
"script": script,
})
output = _run_script(script, base_dir)
_append_comms("IN", "tool_result", {
"name": TOOL_NAME,
"id": block.id,
"output": output,
})
tool_results.append({
"type": "tool_result",
"tool_use_id": block.id,
@@ -360,6 +468,10 @@ def _send_anthropic(md_content: str, user_message: str, base_dir: str) -> str:
"content": tool_results
})
_append_comms("OUT", "tool_result_send", {
"results": [{"tool_use_id": r["tool_use_id"], "content": r["content"]} for r in tool_results],
})
text_parts = [
block.text
for block in response.content

10
config.toml
View File

@@ -17,16 +17,10 @@ paths = [
[screenshots]
base_dir = "C:/Users/Ed/scoop/apps/sharex/current/ShareX/Screenshots/2026-02"
paths = [
"C:/Users/Ed/scoop/apps/sharex/current/ShareX/Screenshots/2026-02/python_2026-02-21_15-24-31.png",
]
paths = []
[discussion]
history = [
"Update the MainContext.md with latest state of codebase. Add support for preserving layout in imgui.",
"[NO response from claude-sonnet-4-6]",
"No code changed but you ran several scripts, whats going on...",
]
history = []
[ai]
provider = "anthropic"

25
dpg_layout.ini
View File

@@ -56,6 +56,31 @@ Size=829,492
Collapsed=0
DockId=0x00000007,0
[Window][###111]
Pos=1578,868
Size=700,440
Collapsed=0
[Window][###126]
Pos=1578,868
Size=700,440
Collapsed=0
[Window][###147]
Pos=1578,868
Size=700,440
Collapsed=0
[Window][###174]
Pos=1578,868
Size=700,440
Collapsed=0
[Window][###207]
Pos=1334,868
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=376,1161 Split=Y Selected=0xEE087978

169
gui.py
View File

@@ -1,4 +1,5 @@
import dearpygui.dearpygui as dpg
# gui.py
import dearpygui.dearpygui as dpg
import tomllib
import tomli_w
import threading
@@ -12,6 +13,9 @@ import shell_runner
CONFIG_PATH = Path("config.toml")
PROVIDERS = ["gemini", "anthropic"]
# Max chars shown inline for a heavy comms field before clamping to a scrollable box
COMMS_CLAMP_CHARS = 300
def load_config() -> dict:
with open(CONFIG_PATH, "rb") as f:
@@ -30,6 +34,76 @@ def hide_tk_root() -> Tk:
return root
# ------------------------------------------------------------------ comms rendering helpers
# Direction -> colour
_DIR_COLORS = {
"OUT": (100, 200, 255), # blue-ish
"IN": (140, 255, 160), # green-ish
}
# Kind -> colour
_KIND_COLORS = {
"request": (255, 220, 100),
"response": (180, 255, 180),
"tool_call": (255, 180, 80),
"tool_result": (180, 220, 255),
"tool_result_send": (200, 180, 255),
}
_HEAVY_KEYS = {"message", "text", "script", "output", "content"}
def _add_comms_field(parent: str, label: str, value: str, heavy: bool):
"""Add a labelled field inside parent. Heavy fields get a clamped input_text box."""
with dpg.group(horizontal=False, parent=parent):
dpg.add_text(f"{label}:", color=(200, 200, 200))
if heavy and len(value) > COMMS_CLAMP_CHARS:
# Show clamped scrollable box
dpg.add_input_text(
default_value=value,
multiline=True,
readonly=True,
width=-1,
height=80,
)
else:
dpg.add_text(value if value else "(empty)", wrap=460)
def _render_comms_entry(parent: str, entry: dict, idx: int):
direction = entry["direction"]
kind = entry["kind"]
ts = entry["ts"]
provider = entry["provider"]
model = entry["model"]
payload = entry["payload"]
dir_color = _DIR_COLORS.get(direction, (220, 220, 220))
kind_color = _KIND_COLORS.get(kind, (220, 220, 220))
with dpg.group(horizontal=False, parent=parent):
# Header row
with dpg.group(horizontal=True):
dpg.add_text(f"#{idx}", color=(160, 160, 160))
dpg.add_text(ts, color=(160, 160, 160))
dpg.add_text(direction, color=dir_color)
dpg.add_text(kind, color=kind_color)
dpg.add_text(f"{provider}/{model}", color=(180, 180, 180))
# Payload fields
for key, val in payload.items():
is_heavy = key in _HEAVY_KEYS
if isinstance(val, (dict, list)):
import json
val_str = json.dumps(val, ensure_ascii=False, indent=2)
else:
val_str = str(val)
_add_comms_field(parent, key, val_str, is_heavy)
dpg.add_separator()
class ConfirmDialog:
"""
Modal confirmation window for a proposed PowerShell script.
@@ -127,8 +201,45 @@ class App:
self._tool_log: list[tuple[str, str]] = []
# Comms log entries queued from background thread for main-thread rendering
self._pending_comms: list[dict] = []
self._pending_comms_lock = threading.Lock()
self._comms_entry_count = 0
ai_client.set_provider(self.current_provider, self.current_model)
ai_client.confirm_and_run_callback = self._confirm_and_run
ai_client.comms_log_callback = self._on_comms_entry
# ---------------------------------------------------------------- comms log
def _on_comms_entry(self, entry: dict):
"""Called from background thread; queue for main thread."""
with self._pending_comms_lock:
self._pending_comms.append(entry)
def _flush_pending_comms(self):
"""Called every frame from the main render loop."""
with self._pending_comms_lock:
entries = self._pending_comms[:]
self._pending_comms.clear()
for entry in entries:
self._comms_entry_count += 1
self._append_comms_entry(entry, self._comms_entry_count)
def _append_comms_entry(self, entry: dict, idx: int):
if not dpg.does_item_exist("comms_scroll"):
return
_render_comms_entry("comms_scroll", entry, idx)
def _rebuild_comms_log(self):
"""Full redraw from ai_client.get_comms_log() — used after clear/reset."""
if not dpg.does_item_exist("comms_scroll"):
return
dpg.delete_item("comms_scroll", children_only=True)
self._comms_entry_count = 0
for entry in ai_client.get_comms_log():
self._comms_entry_count += 1
_render_comms_entry("comms_scroll", entry, self._comms_entry_count)
# ---------------------------------------------------------------- tool execution
@@ -357,8 +468,15 @@ class App:
def cb_reset_session(self):
ai_client.reset_session()
ai_client.clear_comms_log()
self._tool_log.clear()
self._rebuild_tool_log()
# Clear pending queue and counter, then wipe the comms panel
with self._pending_comms_lock:
self._pending_comms.clear()
self._comms_entry_count = 0
if dpg.does_item_exist("comms_scroll"):
dpg.delete_item("comms_scroll", children_only=True)
self._update_status("session reset")
self._update_response("")
@@ -411,6 +529,14 @@ class App:
self._tool_log.clear()
self._rebuild_tool_log()
def cb_clear_comms(self):
ai_client.clear_comms_log()
with self._pending_comms_lock:
self._pending_comms.clear()
self._comms_entry_count = 0
if dpg.does_item_exist("comms_scroll"):
dpg.delete_item("comms_scroll", children_only=True)
# ---------------------------------------------------------------- build ui
def _build_ui(self):
@@ -519,7 +645,7 @@ class App:
tag="win_provider",
pos=(1232, 8),
width=420,
height=280,
height=260,
no_close=True,
):
dpg.add_text("Provider")
@@ -542,13 +668,11 @@ class App:
num_items=6,
callback=self.cb_model_changed,
)
dpg.add_separator()
dpg.add_text("Status: idle", tag="ai_status")
with dpg.window(
label="Message",
tag="win_message",
pos=(1232, 296),
pos=(1232, 276),
width=420,
height=280,
no_close=True,
@@ -568,7 +692,7 @@ class App:
with dpg.window(
label="Response",
tag="win_response",
pos=(1232, 584),
pos=(1232, 564),
width=420,
height=300,
no_close=True,
@@ -584,7 +708,7 @@ class App:
with dpg.window(
label="Tool Calls",
tag="win_tool_log",
pos=(1232, 892),
pos=(1232, 872),
width=420,
height=300,
no_close=True,
@@ -596,6 +720,34 @@ class App:
with dpg.child_window(tag="tool_log_scroll", height=-1, border=False):
pass
# ---- Comms History panel (new) ----
with dpg.window(
label="Comms History",
tag="win_comms",
pos=(1660, 8),
width=520,
height=1164,
no_close=True,
):
# Status line lives here now
with dpg.group(horizontal=True):
dpg.add_text("Status: idle", tag="ai_status", color=(200, 220, 160))
dpg.add_spacer(width=16)
dpg.add_button(label="Clear", callback=self.cb_clear_comms)
dpg.add_separator()
# Colour legend
with dpg.group(horizontal=True):
dpg.add_text("OUT", color=_DIR_COLORS["OUT"])
dpg.add_text("request", color=_KIND_COLORS["request"])
dpg.add_text("tool_call", color=_KIND_COLORS["tool_call"])
dpg.add_spacer(width=8)
dpg.add_text("IN", color=_DIR_COLORS["IN"])
dpg.add_text("response", color=_KIND_COLORS["response"])
dpg.add_text("tool_result", color=_KIND_COLORS["tool_result"])
dpg.add_separator()
with dpg.child_window(tag="comms_scroll", height=-1, border=False, horizontal_scrollbar=True):
pass
def run(self):
dpg.create_context()
dpg.configure_app(docking=True, docking_space=True, init_file="dpg_layout.ini")
@@ -614,6 +766,9 @@ class App:
if dialog is not None:
dialog.show()
# Flush any comms entries queued from background threads
self._flush_pending_comms()
dpg.render_dearpygui_frame()
dpg.save_init_file("dpg_layout.ini")