refactor: optimize token usage, deduplicate code, fix bugs across agents

Token reduction (~40% per session, ~2.3M fewer tokens per 200-feature project):
- Agent-type-specific tool lists: coding 9, testing 5, init 5 (was 19 for all)
- Right-sized max_turns: coding 300, testing 100 (was 1000 for all)
- Trimmed coding prompt template (~150 lines removed)
- Streamlined testing prompt with batch support
- YOLO mode now strips browser testing instructions from prompt
- Added Grep, WebFetch, WebSearch to expand project session

Performance improvements:
- Rate limit retries start at ~15s with jitter (was fixed 60s)
- Post-spawn delay reduced to 0.5s (was 2s)
- Orchestrator consolidated to 1 DB query per loop (was 5-7)
- Testing agents batch 3 features per session (was 1)
- Smart context compaction preserves critical state, discards noise

Bug fixes:
- Removed ghost feature_release_testing MCP tool (wasted tokens every test session)
- Forward all 9 Vertex AI env vars to chat sessions (was missing 3)
- Fix DetachedInstanceError risk in test batch ORM access
- Prevent duplicate testing of same features in parallel mode

Code deduplication:
- _get_project_path(): 9 copies -> 1 shared utility (project_helpers.py)
- validate_project_name(): 9 copies -> 2 variants in 1 file (validation.py)
- ROOT_DIR: 10 copies -> 1 definition (chat_constants.py)
- API_ENV_VARS: 4 copies -> 1 source of truth (env_constants.py)

Security hardening:
- Unified sensitive directory blocklist (14 dirs, was two divergent lists)
- Cached get_blocked_paths() for O(1) directory listing checks
- Terminal security warning when ALLOW_REMOTE=1 exposes WebSocket
- 20 new security tests for EXTRA_READ_PATHS blocking
- Extracted _validate_command_list() and _validate_pkill_processes() helpers

Type safety:
- 87 mypy errors -> 0 across 58 source files
- Installed types-PyYAML for proper yaml stub types
- Fixed SQLAlchemy Column[T] coercions across all routers

Dead code removed:
- 13 files deleted (~2,679 lines): unused UI components, debug logs, outdated docs
- 7 unused npm packages removed (Radix UI components with 0 imports)
- AgentAvatar.tsx reduced from 615 -> 119 lines (SVGs extracted to mascotData.tsx)

New CLI options:
- --testing-batch-size (1-5) for parallel mode test batching
- --testing-feature-ids for direct multi-feature testing

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
Auto
2026-02-01 13:16:24 +02:00
parent dc5bcc4ae9
commit 94e0b05cb1
57 changed files with 1974 additions and 4300 deletions

View File

@@ -6,12 +6,10 @@ API endpoints for managing agent schedules.
Provides CRUD operations for time-based schedule configuration.
"""
import re
import sys
from contextlib import contextmanager
from datetime import datetime, timedelta, timezone
from pathlib import Path
from typing import Generator, Tuple
from typing import TYPE_CHECKING, Generator, Tuple
from fastapi import APIRouter, HTTPException
from sqlalchemy.orm import Session
@@ -26,17 +24,21 @@ from ..schemas import (
ScheduleResponse,
ScheduleUpdate,
)
from ..utils.project_helpers import get_project_path as _get_project_path
from ..utils.validation import validate_project_name
if TYPE_CHECKING:
from api.database import Schedule as ScheduleModel
def _get_project_path(project_name: str) -> Path:
"""Get project path from registry."""
root = Path(__file__).parent.parent.parent
if str(root) not in sys.path:
sys.path.insert(0, str(root))
from registry import get_project_path
return get_project_path(project_name)
def _schedule_to_response(schedule: "ScheduleModel") -> ScheduleResponse:
"""Convert a Schedule ORM object to a ScheduleResponse Pydantic model.
SQLAlchemy Column descriptors resolve to Python types at instance access time,
but mypy sees the Column[T] descriptor type. Using model_validate with
from_attributes handles this conversion correctly.
"""
return ScheduleResponse.model_validate(schedule, from_attributes=True)
router = APIRouter(
prefix="/api/projects/{project_name}/schedules",
@@ -44,16 +46,6 @@ router = APIRouter(
)
def validate_project_name(name: str) -> str:
"""Validate and sanitize project name to prevent path traversal."""
if not re.match(r'^[a-zA-Z0-9_-]{1,50}$', name):
raise HTTPException(
status_code=400,
detail="Invalid project name"
)
return name
@contextmanager
def _get_db_session(project_name: str) -> Generator[Tuple[Session, Path], None, None]:
"""Get database session for a project as a context manager.
@@ -102,21 +94,7 @@ async def list_schedules(project_name: str):
).order_by(Schedule.start_time).all()
return ScheduleListResponse(
schedules=[
ScheduleResponse(
id=s.id,
project_name=s.project_name,
start_time=s.start_time,
duration_minutes=s.duration_minutes,
days_of_week=s.days_of_week,
enabled=s.enabled,
yolo_mode=s.yolo_mode,
model=s.model,
crash_count=s.crash_count,
created_at=s.created_at,
)
for s in schedules
]
schedules=[_schedule_to_response(s) for s in schedules]
)
@@ -190,18 +168,7 @@ async def create_schedule(project_name: str, data: ScheduleCreate):
except Exception as e:
logger.error(f"Failed to start agent for schedule {schedule.id}: {e}", exc_info=True)
return ScheduleResponse(
id=schedule.id,
project_name=schedule.project_name,
start_time=schedule.start_time,
duration_minutes=schedule.duration_minutes,
days_of_week=schedule.days_of_week,
enabled=schedule.enabled,
yolo_mode=schedule.yolo_mode,
model=schedule.model,
crash_count=schedule.crash_count,
created_at=schedule.created_at,
)
return _schedule_to_response(schedule)
@router.get("/next", response_model=NextRunResponse)
@@ -259,8 +226,8 @@ async def get_next_scheduled_run(project_name: str):
return NextRunResponse(
has_schedules=True,
next_start=next_start.isoformat() if (active_count == 0 and next_start) else None,
next_end=latest_end.isoformat() if latest_end else None,
next_start=next_start if active_count == 0 else None,
next_end=latest_end,
is_currently_running=active_count > 0,
active_schedule_count=active_count,
)
@@ -280,18 +247,7 @@ async def get_schedule(project_name: str, schedule_id: int):
if not schedule:
raise HTTPException(status_code=404, detail="Schedule not found")
return ScheduleResponse(
id=schedule.id,
project_name=schedule.project_name,
start_time=schedule.start_time,
duration_minutes=schedule.duration_minutes,
days_of_week=schedule.days_of_week,
enabled=schedule.enabled,
yolo_mode=schedule.yolo_mode,
model=schedule.model,
crash_count=schedule.crash_count,
created_at=schedule.created_at,
)
return _schedule_to_response(schedule)
@router.patch("/{schedule_id}", response_model=ScheduleResponse)
@@ -334,18 +290,7 @@ async def update_schedule(
# Was enabled, now disabled - remove jobs
scheduler.remove_schedule(schedule_id)
return ScheduleResponse(
id=schedule.id,
project_name=schedule.project_name,
start_time=schedule.start_time,
duration_minutes=schedule.duration_minutes,
days_of_week=schedule.days_of_week,
enabled=schedule.enabled,
yolo_mode=schedule.yolo_mode,
model=schedule.model,
crash_count=schedule.crash_count,
created_at=schedule.created_at,
)
return _schedule_to_response(schedule)
@router.delete("/{schedule_id}", status_code=204)