ISSUE_TRIAGE_POLICY

This repository uses an AI-assisted issue triage workflow for Nexus Terminal.

Goals

• Reduce repeated support load for deployment and configuration issues
• Identify whether a problem belongs to frontend, backend, remote-gateway, or docs
• Provide actionable troubleshooting steps for operators
• Propose implementation plans for likely product defects
• Require explicit maintainer approval before code changes

Product context

Nexus Terminal is a web-based remote access platform covering:

• SSH / SFTP
• RDP / VNC
• Docker-based deployment
• reverse proxy deployment
• authentication, captcha, and 2FA
• audit and security-related controls

Because of this, many reported issues are caused by:

• reverse proxy and websocket forwarding
• environment variables
• Docker deployment
• CORS
• auth/security configuration
• protocol-specific remote connection setup

Classification

Every issue should be classified as one of:

• deployment_or_proxy_issue
• auth_or_security_config_issue
• connection_protocol_issue
• frontend_behavior_issue
• backend_logic_issue
• docs_gap
• feature_request
• unclear

Decision rules

deployment_or_proxy_issue

Use this when the issue is likely caused by:

• nginx or reverse proxy configuration
• websocket upgrade handling
• Docker / compose deployment
• environment variables
• port exposure
• TLS / forwarded headers
• CORS or browser-origin behavior

auth_or_security_config_issue

Use this when the issue is likely caused by:

• login flow configuration
• captcha configuration
• 2FA behavior
• IP whitelist / blacklist
• security restriction misunderstanding

connection_protocol_issue

Use this when the issue is mainly related to:

• SSH
• SFTP
• RDP
• VNC
• keyboard-interactive auth
• remote desktop gateway behavior
• remote session stability

frontend_behavior_issue

Use this when the issue is mainly:

• UI rendering
• theme/layout
• terminal display or interaction
• file manager UX
• client-side caching/state
• browser-only failures

backend_logic_issue

Use this only when repository context strongly suggests a real backend or gateway defect.

docs_gap

Use this when the likely answer exists but the documentation is missing, insufficient, or hard to find.

feature_request

Use this when the report is asking for a new capability rather than describing a broken one.

unclear

Use this when the report lacks enough detail for responsible diagnosis.

Automation policy

May auto-reply

• deployment_or_proxy_issue
• auth_or_security_config_issue
• connection_protocol_issue caused by configuration
• docs_gap
• unclear

Must not auto-implement

• auth/session logic
• security controls
• audit logic
• credential handling
• access control
• IP ban/allow logic
• low-confidence diagnoses
• issues without a plausible reproduction path

May propose implementation plan

• backend_logic_issue
• frontend_behavior_issue
• connection_protocol_issue with clear defect evidence
• feature_request

May implement only after maintainer approval

A maintainer must comment:

/codex implement

Required follow-up information

When more information is required, ask only for:

• Nexus Terminal version
• deployment method (Docker / source / other)
• reverse proxy in use (nginx / caddy / none / other)
• browser and OS if UI-related
• exact protocol affected (SSH / SFTP / RDP / VNC)
• exact steps to reproduce
• exact error output
• sanitized config snippet if relevant
• whether the issue reproduces behind direct access vs reverse proxy

Default doc lookup order

1. README.md
2. docs/deployment/nginx.md
3. docs/configuration/docker.md
4. docs/configuration/cors.md
5. other docs/**
6. matching workspace source and tests