Proposal: fix architecture-guard — use correct issue labels (Type/Refactoring) instead of Automation Tracking when filing refactoring issues #5562

New issue

Open

opened 2026-04-09 07:33:12 +00:00 by HAL9000 · 0 comments

HAL9000 commented

2026-04-09 07:33:12 +00:00

Owner

Agent Improvement Proposal

Pattern Detected

Type: workflow_fix
Affected Agent: architecture-guard
Evidence: Issue #5536 ([AUTO-GUARD] Refactor duplicate TUI widget base functions) was filed by architecture-guard with the "Automation Tracking" label applied. This is incorrect — "Automation Tracking" is reserved for agent health/status tracking issues, not for refactoring findings. The architecture-guard is conflating its tracking issue format with its refactoring issue format, causing refactoring findings to appear in automation tracking queries and polluting the tracking system.

Specifically, the architecture-guard's tracking functions apply "Automation Tracking" to ALL issues it creates, including refactoring findings that should use "Type/Refactoring" labels. The tracking functions in the agent definition only apply "Automation Tracking" to the cycle status reports, but the agent appears to be applying this label to refactoring issues as well.

Proposed Change

Modify the architecture-guard agent definition to explicitly clarify that:

Tracking/status issues (e.g., [AUTO-GUARD] Code Analysis Report (Cycle N)) should use "Automation Tracking" + "Type/Automation" + "State/In Progress" + "Priority/Medium"
Refactoring finding issues (e.g., [AUTO-GUARD] Refactor duplicate TUI widget base functions) should use "Type/Refactoring" + appropriate priority + "State/Unverified" — NOT "Automation Tracking"

Add explicit guidance in the agent's "What to Check" section and issue creation instructions to use the correct label set for each issue type.

Expected Impact

Automation tracking queries will return only actual automation status issues
Refactoring issues will be properly labeled and discoverable by implementation workers
The backlog groomer will correctly categorize refactoring issues
Reduces noise in the automation tracking system

Risk Assessment

Low risk: This is a label correction, not a behavioral change
The architecture-guard will continue to file the same issues, just with correct labels
No impact on the agent's scanning or analysis capabilities

This is a proposal from the agent evolver. A human must approve this issue before the change will be implemented. To approve: remove the needs feedback label, add State/Verified, or comment with approval.

Automated by CleverAgents Bot
Supervisor: Agent Evolver | Agent: agent-evolver

## Agent Improvement Proposal ### Pattern Detected **Type**: workflow_fix **Affected Agent**: architecture-guard **Evidence**: Issue #5536 (`[AUTO-GUARD] Refactor duplicate TUI widget base functions`) was filed by architecture-guard with the "Automation Tracking" label applied. This is incorrect — "Automation Tracking" is reserved for agent health/status tracking issues, not for refactoring findings. The architecture-guard is conflating its tracking issue format with its refactoring issue format, causing refactoring findings to appear in automation tracking queries and polluting the tracking system. Specifically, the architecture-guard's tracking functions apply "Automation Tracking" to ALL issues it creates, including refactoring findings that should use "Type/Refactoring" labels. The tracking functions in the agent definition only apply "Automation Tracking" to the cycle status reports, but the agent appears to be applying this label to refactoring issues as well. ### Proposed Change Modify the architecture-guard agent definition to explicitly clarify that: 1. **Tracking/status issues** (e.g., `[AUTO-GUARD] Code Analysis Report (Cycle N)`) should use "Automation Tracking" + "Type/Automation" + "State/In Progress" + "Priority/Medium" 2. **Refactoring finding issues** (e.g., `[AUTO-GUARD] Refactor duplicate TUI widget base functions`) should use "Type/Refactoring" + appropriate priority + "State/Unverified" — NOT "Automation Tracking" Add explicit guidance in the agent's "What to Check" section and issue creation instructions to use the correct label set for each issue type. ### Expected Impact - Automation tracking queries will return only actual automation status issues - Refactoring issues will be properly labeled and discoverable by implementation workers - The backlog groomer will correctly categorize refactoring issues - Reduces noise in the automation tracking system ### Risk Assessment - **Low risk**: This is a label correction, not a behavioral change - The architecture-guard will continue to file the same issues, just with correct labels - No impact on the agent's scanning or analysis capabilities --- *This is a proposal from the agent evolver. A human must approve this issue before the change will be implemented. To approve: remove the `needs feedback` label, add `State/Verified`, or comment with approval.* --- **Automated by CleverAgents Bot** Supervisor: Agent Evolver | Agent: agent-evolver

Rows
Columns