tilt-qa.md•4.95 kB
# Tilt QA Demonstration and Bug Report
Run a comprehensive quality assurance session for the Tilt MCP integration, testing all tools and identifying bugs.
## Instructions
Execute the following QA workflow systematically, documenting results at each step:
### Phase 1: Discovery and Initial Status
1. **Check Overall Status**
- Use `tilt_status` to get system-wide status
- Document: resources count, overall state, any warnings
- ❌ **Bug check**: Missing data? Incorrect formatting? Errors?
### Phase 2: Resource Operations
3. **List Resources**
- Use `tilt_get_resources` with different filters:
- All resources (default)
- Filter by status (ok, error, pending, building)
- With verbose output
- With pagination (limit/offset)
- Document: resources found, their states, response times
- ❌ **Bug check**: Pagination issues? Filtering broken? Missing fields?
4. **Describe Resources**
- Use `tilt_describe_resource` on 2-3 different resources
- Test with different resource types if available
- Document: detail level, completeness, formatting
- ❌ **Bug check**: Invalid resource names handling? Incomplete data? Errors?
### Phase 3: Logs and Monitoring
5. **Read Logs**
- Use `tilt_logs` (plain-text output) with various options:
- Default tail (100 lines)
- Different tail line counts (50, 200)
- Filter by source (build, runtime, all)
- Filter by level (warn, error) — note: filters Tilt system messages only
- Client-side search:
- Substring match (case-sensitive and case-insensitive)
- Regex with flags (e.g., `im`), ensure invalid regex throws clear error
- Document: log output, filtering effectiveness, search behavior, performance
- ❌ **Bug check**: Missing logs? Incorrect filtering? Search not applied? Timeout issues? Errors not descriptive for bad regex?
### Phase 4: Resource Control
6. **Trigger Updates**
- Use `tilt_trigger` on a resource (choose one safely)
- Document: trigger response, resource state change
- ❌ **Bug check**: Trigger fails? No feedback? State inconsistency?
7. **Enable/Disable Resources**
- Use `tilt_disable` on a resource
- Verify state changed using `tilt_describe_resource`
- Use `tilt_enable` to re-enable
- Verify state restored
- Document: state transitions, response times
- ❌ **Bug check**: State not changing? Errors? Inconsistent behavior?
### Phase 5: Advanced Features
8. **Wait Operations**
- Use `tilt_wait` on one or more resources
- Test with different timeouts
- Document: wait behavior, timeout handling
- ❌ **Bug check**: Timeout not respected? Hang indefinitely? Incorrect ready state?
9. **Tiltfile Arguments**
- Use `tilt_args` to view current arguments
- Set new arguments if safe to do so
- Clear arguments
- Document: argument handling, effects on resources
- ❌ **Bug check**: Args not applied? Clear fails? Incorrect parsing?
### Phase 6: Error Handling and Edge Cases
10. **Test Error Conditions**
- Call `tilt_describe_resource` with invalid resource name
- Call `tilt_logs` with invalid resource name
- Call `tilt_logs` with invalid search regex and confirm explicit failure
- Test with Tilt not running (if possible to test safely)
- Document: error messages, error handling quality
- ❌ **Bug check**: Poor error messages? Crashes? Undefined behavior?
## Bug Report Format
After completing all phases, create a structured bug report with the following sections:
### Executive Summary
- Total tools tested: X
- Tools working correctly: X
- Bugs found: X
- Critical issues: X
- Severity breakdown
### Critical Bugs (Blocks core functionality)
For each bug:
- **Bug ID**: TILT-XXX
- **Tool**: tool_name
- **Summary**: One-line description
- **Steps to Reproduce**: Numbered steps
- **Expected Behavior**: What should happen
- **Actual Behavior**: What actually happens
- **Impact**: How this affects users
- **Severity**: Critical/High/Medium/Low
### Medium/Low Priority Bugs
Same format as above
### Quality Issues (Not bugs but improvements needed)
- Inconsistent error messages
- Missing validation
- Poor user experience
- Documentation gaps
### Successful Test Cases
Document what works correctly to establish baseline
### Recommendations
1. Priority fixes (in order)
2. Testing improvements needed
3. Documentation updates required
## Success Criteria
- ✅ All 9 Tilt MCP tools tested
- ✅ Error conditions tested for each tool
- ✅ Performance documented
- ✅ Bug report generated with severity levels
- ✅ Reproduction steps for each bug
## Notes
- Run this QA session against a **non-production** Tilt instance
- Document exact versions of tools, Tilt CLI, and dependencies
- Take screenshots or save logs for critical bugs
- Test on the actual environment where bugs occur
- Be systematic - complete each phase before moving to the next