Commit Graph

2 Commits

Author SHA1 Message Date
Claude
838927f316 Fix webhook error handling with reliable error display
- API now properly detects N8N errors in 200 responses (OpenAI quota, etc.)
- Returns 400 status with simple "Agent unavailable" message for webhook failures
- Frontend shows errors in results container instead of disappearing toast notifications
- Users now see clear, persistent error messages when agents fail
- Keeps success toasts working, only fixes error display

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-15 12:38:43 +05:30
Claude
4b64cb2d29 🏗️ Refactor agents views into focused modules for better code organization
**View Separation Completed:**
- **agents/api_views.py** - REST API endpoints (execute_agent, execution_list/detail)
- **agents/chat_views.py** - Chat session management and message handling
- **agents/web_views.py** - Web interface views (marketplace, agent detail pages)
- **agents/direct_access_views.py** - External form integration handlers
- **agents/utils.py** - Utility functions (webhook validation, message formatting, AgentCompat class)

**Benefits:**
 Better code navigation and maintainability (1,433 lines → 5 focused modules)
 Clear separation of concerns (API vs web vs chat vs external forms)
 Easier to add new agent types in the future
 Follows Django best practices
 Maintains full backwards compatibility through main views.py imports

**Architecture Status:**
- All 8 agents working perfectly (4 webhook + 4 direct access)
- File-based agent system with production-ready caching
- Wallet balance updates immediately after execution
- PDF analyzer showing properly formatted results
- Job posting generator with clean formatting
- "Explore Other Agents" button functioning correctly
- N8N webhook communication working seamlessly

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-14 22:36:36 +05:30