Automatically detects technical errors and translates them into plain-language explanations adapted to user expertise level. Categorizes severity, provides context-aware recovery guidance, and offers prioritized action steps. Transforms confusing error codes into constructive problem-solving opportunities.
RCP-001-001-010-ERROR-TRANS
The CRAFT Recipe
A: The recipe activates when the AI encounters complex
technical errors that would benefit from translation
before presenting to the user. Q: What if the error is already clear?
A: Simple errors with complexity score below 4 are
passed through without translation. The recipe only
adds value when translation improves clarity. Q: How does auto-detect expertise work?
A: The AI analyzes conversation history for technical
terms, code snippets, and question complexity to
determine appropriate tone adaptation. Q: Can I override the severity assessment?
A: Yes, you can specify severity context when providing
the error message for more accurate categorization. Q: Does this recipe integrate with other recipes?
A: Yes, it works with EXPECT-SET for communicating
limitations and PROGRESS-TRACK for error recovery
status updates. Â EXAMPLE 1: Technical Error for Non-Technical User Input:
error_message: “TypeError: Cannot read property
‘map’ of undefined at UserList.render”
error_context: “I was trying to view my contacts”
user_expertise: non-technical Output:
#AI->H::Note: (WARNING – Display Issue) The contact list could not load because it
received empty data instead of your contacts. In simple terms: Imagine opening a filing cabinet
expecting folders but finding it empty. The system
looked for your contacts but found nothing there. #AI->H::Question: (Would you like me to explain
how to fix this?) EXAMPLE 2: API Error for Technical User Input:
error_message: “HTTP 429 Too Many Requests –
Rate limit exceeded for API endpoint”
error_context: “Running automated data sync”
user_expertise: technical Output:
#AI->H::Note: (WARNING – Rate Limiting) API rate limit exceeded. Your automated sync is
sending requests faster than the endpoint allows. How to fix this:
1. Add delay between requests (recommended
500ms minimum)
2. Implement exponential backoff on retries
3. Check API documentation for rate limits
4. Consider batching multiple operations #AI->H::Note: (This can be resolved by adjusting
your sync configuration) EXAMPLE 3: Critical System Error Input:
error_message: “ECONNREFUSED 127.0.0.1:5432 –
Connection refused”
error_context: “Application won’t start”
user_expertise: auto-detect Output:
#AI->H::Caution: (CRITICAL – Database Unavailable) The application cannot connect to the database
server. This prevents the application from
starting or functioning. In simple terms: The application is trying to
talk to the database, but nobody is answering.
The database service may not be running. How to fix this:
1. Quick fix: Check if database service is
running and restart if needed
2. Workaround: Verify connection settings
match database configuration
3. Prevention: Set up monitoring to alert
when database becomes unavailable #AI->H::Note: (This needs assistance from admin
or developer if database service restart does
not resolve the issue)
