[Enhancement] /debug-report: Add user-reported issue mode #139

New Issue

lmiranda · 2026-01-24T17:26:28Z

lmiranda commented

2026-01-24 17:26:28 +00:00

Problem

Currently /debug-report is focused on running automated MCP tool tests and reporting pass/fail results. However, users often need to report issues they experience while using plugin commands - not issues caught by automated tests.

Example scenario from #137:

User tried to use virt_update_vm to update a VM's description
Tool schema only exposed id parameter, not description
This wasn't a "test failure" - the tool technically worked, but was missing functionality
User had to manually create the diagnostic issue

Current Flow

Runs 5 automated MCP tool tests
Reports pass/fail results
Creates issue based on test failures

Proposed Enhancement

Add a user feedback mode to /debug-report that:

Step 1: Ask for issue type

What would you like to report?

[ ] Automated diagnostics - Run MCP tool tests
[ ] User-reported issue - Describe a problem you experienced

Step 2 (if user-reported): Gather context

What were you trying to do?
- Which command/tool were you using?
- What was the goal?
What went wrong?
- Error message (if any)
- Unexpected behavior
- Missing functionality
Expected vs Actual
- What did you expect to happen?
- What actually happened?
Reproduction steps (optional)
- How can this be reproduced?

Step 3: Auto-capture context

Still capture automatically:

Git remote / repository
Current branch
Working directory
Relevant tool schemas (if identifiable)

Step 4: Create structured issue

Generate issue with:

User's description of the problem
Auto-captured context
Category tags based on problem type
Suggested investigation areas

Benefits

Users can report UX issues, missing features, documentation gaps
Not limited to "test failures"
Still captures full context automatically
Creates well-structured issues for investigation

Implementation Notes

Add a question at the start of /debug-report to select mode
Reuse existing context-gathering logic
Keep automated test mode as default for backwards compatibility
User feedback mode could skip the automated tests entirely

Issue #137 was a user-reported issue that had to be manually documented
The automated tests in that report (404 errors) were irrelevant to the actual problem

## Problem Currently `/debug-report` is focused on running automated MCP tool tests and reporting pass/fail results. However, users often need to report issues they **experience while using plugin commands** - not issues caught by automated tests. **Example scenario from #137:** - User tried to use `virt_update_vm` to update a VM's description - Tool schema only exposed `id` parameter, not `description` - This wasn't a "test failure" - the tool technically worked, but was missing functionality - User had to manually create the diagnostic issue ## Current Flow 1. Runs 5 automated MCP tool tests 2. Reports pass/fail results 3. Creates issue based on test failures ## Proposed Enhancement Add a **user feedback mode** to `/debug-report` that: ### Step 1: Ask for issue type ``` What would you like to report? [ ] Automated diagnostics - Run MCP tool tests [ ] User-reported issue - Describe a problem you experienced ``` ### Step 2 (if user-reported): Gather context 1. **What were you trying to do?** - Which command/tool were you using? - What was the goal? 2. **What went wrong?** - Error message (if any) - Unexpected behavior - Missing functionality 3. **Expected vs Actual** - What did you expect to happen? - What actually happened? 4. **Reproduction steps** (optional) - How can this be reproduced? ### Step 3: Auto-capture context Still capture automatically: - Git remote / repository - Current branch - Working directory - Relevant tool schemas (if identifiable) ### Step 4: Create structured issue Generate issue with: - User's description of the problem - Auto-captured context - Category tags based on problem type - Suggested investigation areas ## Benefits - Users can report UX issues, missing features, documentation gaps - Not limited to "test failures" - Still captures full context automatically - Creates well-structured issues for investigation ## Implementation Notes - Add a question at the start of `/debug-report` to select mode - Reuse existing context-gathering logic - Keep automated test mode as default for backwards compatibility - User feedback mode could skip the automated tests entirely ## Related - Issue #137 was a user-reported issue that had to be manually documented - The automated tests in that report (404 errors) were irrelevant to the actual problem

lmiranda referenced this issue from a commit

2026-01-24 17:30:46 +00:00

feat(projman): add user-reported issue mode to /debug-report

lmiranda referenced a pull request that will close this issue

2026-01-24 17:31:07 +00:00

feat(projman): add user-reported issue mode to /debug-report #141

lmiranda closed this issue

2026-01-24 17:36:16 +00:00

Sign in to join this conversation.