25 lines
1.9 KiB
Markdown
25 lines
1.9 KiB
Markdown
# TODO
|
|
|
|
## P0
|
|
- [Bug] Enforce single active desktop-control run (or a strict queue) so concurrent jobs cannot fight over the same mouse/keyboard/screen session.
|
|
- [Bug] Fix run artifact collisions in `setup_artifacts()` (`run_id` is second-granularity, so two jobs in the same second can share/overwrite the same directory).
|
|
- [Bug] Remove global logger handler clobbering in `setup_logger()` (`logging.getLogger("screenjob").handlers.clear()` breaks concurrent runs and can redirect logs to the wrong file).
|
|
- [Bug] More consistent clicks and more uses of enhance images.
|
|
|
|
## P1
|
|
- [Bug] Validate `disabled_tools` against an allowlist and disallow disabling critical completion flow (`task_complete`) to avoid guaranteed step-limit failures.
|
|
- [Bug] Improve `execute_command` cancellation/timeout handling to terminate full process trees, not only the parent shell process.
|
|
- [Bug] Reduce API/UI token leakage risk by moving away from query-string token usage for websocket/artifact access where possible.
|
|
- [Idea] Add per-token rate limiting and request size limits (objective length + payload bounds) for API hardening.
|
|
|
|
## P2
|
|
- [Bug] Fix UI event style mapping mismatch (`tool_called` events are emitted, but UI color map expects `tool_call`).
|
|
- [Idea] Reduce monitoring UI backend load by throttling websocket-triggered refreshes and avoiding full job/event re-fetch on every event.
|
|
- [Idea] Add cursor-based pagination for jobs/events instead of large fixed limits.
|
|
- [Idea] Support offline/self-hosted UI assets (bundle Tailwind instead of CDN dependency).
|
|
- [Idea] Add retention controls/pruning for old runs, screenshots, and DB rows.
|
|
|
|
## P3
|
|
- [Idea] Add Replay Mode; Ability to replay a session by reconstructing the screen from screenshots and overlaying tool calls and click and type events.
|
|
- [Idea] Add lightweight analytics dashboards (success rate by objective category, avg steps/cost over time).
|