Agent Skills

.github/skills/api-development/SKILL.md

@@ -0,0 +1,58 @@
+---
+name: api-development
+description: Develop and extend NetAlertX REST API endpoints. Use this when asked to create endpoint, add API route, implement API, or modify API responses.
+---
+
+# API Development
+
+## Entry Point
+
+Flask app: `server/api_server/api_server_start.py`
+
+## Existing Routes
+
+- `/device/<mac>` - Single device operations
+- `/devices` - Device list
+- `/devices/export/{csv,json}` - Export devices
+- `/devices/import` - Import devices
+- `/devices/totals` - Device counts
+- `/devices/by-status` - Devices grouped by status
+- `/nettools` - Network utilities
+- `/events` - Event log
+- `/sessions` - Session management
+- `/dbquery` - Database queries
+- `/metrics` - Prometheus metrics
+- `/sync` - Synchronization
+
+## Authorization
+
+All routes require header:
+
+```
+Authorization: Bearer <API_TOKEN>
+```
+
+Retrieve token via `get_setting_value('API_TOKEN')`.
+
+## Response Contract
+
+**MANDATORY:** All responses must include `"success": true|false`
+
+On failure, include error message:
+
+```python
+return {"success": False, "error": "Description of what went wrong"}
+```
+
+On success:
+
+```python
+return {"success": True, "data": result}
+```
+
+## Adding New Endpoints
+
+1. Add route in `server/api_server/` directory
+2. Follow authorization pattern
+3. Return proper response contract
+4. Update UI to read/write JSON cache (don't bypass pipeline)

.github/skills/authentication/SKILL.md

@@ -0,0 +1,60 @@
+---
+name: netalertx-authentication-tokens
+description: Manage and troubleshoot API tokens and authentication-related secrets. Use this when you need to find, rotate, verify, or debug authentication issues (401/403) in NetAlertX.
+---
+
+# Authentication
+
+## Purpose ✅
+Explain how to locate, validate, rotate, and troubleshoot API tokens and related authentication settings used by NetAlertX.
+
+## Pre-Flight Check (MANDATORY) ⚠️
+1. Ensure the backend is running (use devcontainer services or `ps`/systemd checks).
+2. Verify the `API_TOKEN` setting can be read with Python (see below).
+3. If a token-related error occurs, gather logs (`/tmp/log/app.log`, nginx logs) before changing secrets.
+
+## Retrieve the API token (Python — preferred) 🐍
+Always use Python helpers to read secrets to avoid accidental exposure in shells or logs:
+
+```python
+from helper import get_setting_value
+token = get_setting_value("API_TOKEN")
+```
+
+If you must inspect from a running container (read-only), use:
+
+```bash
+docker exec <CONTAINER_ID> python3 -c "from helper import get_setting_value; print(get_setting_value('API_TOKEN'))"
+```
+
+You can also check the runtime config file:
+
+```bash
+docker exec <CONTAINER_ID> grep API_TOKEN /data/config/app.conf
+```
+
+## Rotate / Generate a new token 🔁
+- Preferred: Use the web UI (Settings / System) and click **Generate** for the `API_TOKEN` field — this updates the value safely and immediately.
+- Manual: Edit `/data/config/app.conf` and restart the backend if required (use the existing devcontainer service tasks).
+- After rotation: verify the value with `get_setting_value('API_TOKEN')` and update any clients or sync nodes to use the new token.
+
+## Troubleshooting 401 / 403 Errors 🔍
+1. Confirm backend is running and reachable.
+2. Confirm `get_setting_value('API_TOKEN')` returns a non-empty value.
+3. Ensure client requests send the header exactly: `Authorization: Bearer <API_TOKEN>`.
+4. Check `/tmp/log/app.log` and plugin logs (e.g., sync plugin) for "Incorrect API Token" messages.
+5. If using multiple nodes, ensure the token matches across nodes for sync operations.
+6. If token appears missing or incorrect, rotate via UI or update `app.conf` and re-verify.
+
+## Best Practices & Security 🔐
+- Never commit tokens to source control or paste them in public issues. Redact tokens when sharing logs.
+- Rotate tokens when a secret leak is suspected or per your security policy.
+- Use `get_setting_value()` in tests and scripts — do not hardcode secrets.
+
+## Related Skills & Docs 📚
+- `testing-workflow` — how to use `API_TOKEN` in tests
+- `settings-management` — where settings live and how they are managed
+- Docs: `docs/API.md`, `docs/API_OLD.md`, `docs/API_SSE.md`
+
+---
+_Last updated: 2026-01-23_

.github/skills/code-standards/SKILL.md

@@ -0,0 +1,65 @@
+---
+name: netalertx-code-standards
+description: NetAlertX coding standards and conventions. Use this when writing code, reviewing code, or implementing features.
+---
+
+# Code Standards
+
+## File Length
+
+Keep code files under 500 lines. Split larger files into modules.
+
+## DRY Principle
+
+Do not re-implement functionality. Reuse existing methods or refactor to create shared methods.
+
+## Database Access
+
+- Never access DB directly from application layers
+- Use `server/db/db_helper.py` functions (e.g., `get_table_json`)
+- Implement new functionality in handlers (e.g., `DeviceInstance` in `server/models/device_instance.py`)
+
+## MAC Address Handling
+
+Always validate and normalize MACs before DB writes:
+
+```python
+from plugin_helper import normalize_mac
+
+mac = normalize_mac(raw_mac)
+```
+
+## Subprocess Safety
+
+**MANDATORY:** All subprocess calls must set explicit timeouts.
+
+```python
+result = subprocess.run(cmd, timeout=60)  # Minimum 60s
+```
+
+Nested subprocess calls need their own timeout—outer timeout won't save you.
+
+## Time Utilities
+
+```python
+from utils.datetime_utils import timeNowDB
+
+timestamp = timeNowDB()
+```
+
+## String Sanitization
+
+Use sanitizers from `server/helper.py` before storing user input.
+
+## Devcontainer Constraints
+
+- Never `chmod` or `chown` during operations
+- Everything is already writable
+- If permissions needed, fix `.devcontainer/scripts/setup.sh`
+
+## Path Hygiene
+
+- Use environment variables for runtime paths
+- `/data` for persistent config/db
+- `/tmp` for runtime logs/api/nginx state
+- Never hardcode `/data/db` or use relative paths

.github/skills/database-reset/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: reset-netalertx-database
+description: Wipe and regenerate the NetAlertX database and config. Use this when asked to reset database, wipe db, fresh database, clean slate, or start fresh.
+---
+
+# Database Reset
+
+Completely wipes devcontainer database and config, then regenerates from scratch.
+
+## Command
+
+```bash
+killall 'python3' || true
+sleep 1
+rm -rf /data/db/* /data/config/*
+bash /entrypoint.d/15-first-run-config.sh
+bash /entrypoint.d/20-first-run-db.sh
+```
+
+## What This Does
+
+1. Kills backend to release database locks
+2. Deletes all files in `/data/db/` and `/data/config/`
+3. Runs first-run config provisioning
+4. Runs first-run database initialization
+
+## After Reset
+
+Run the startup script to restart services:
+
+```bash
+/workspaces/NetAlertX/.devcontainer/scripts/setup.sh
+```
+
+## Database Location
+
+- Runtime: `/data/db/app.db` (SQLite)
+- Config: `/data/config/app.conf`

.github/skills/devcontainer-configs/SKILL.md

@@ -0,0 +1,28 @@
+---
+name: netalertx-devcontainer-configs
+description: Generate devcontainer configuration files. Use this when asked to generate devcontainer configs, update devcontainer template, or regenerate devcontainer.
+---
+
+# Devcontainer Config Generation
+
+Generates devcontainer configs from the template. Must be run after changes to devcontainer configuration.
+
+## Command
+
+```bash
+/workspaces/NetAlertX/.devcontainer/scripts/generate-configs.sh
+```
+
+## What It Does
+
+Combines and merges template configurations into the final config used by VS Code.
+
+## When to Run
+
+- After modifying `.devcontainer/` template files
+- After changing devcontainer features or settings
+- Before committing devcontainer changes
+
+## Note
+
+This affects only the devcontainer configuration. It has no bearing on the production or test Docker image.

.github/skills/devcontainer-services/SKILL.md

@@ -0,0 +1,50 @@
+---
+name: restarting-netalertx-services
+description: Control NetAlertX services inside the devcontainer. Use this when asked to start backend, start frontend, start nginx, start php-fpm, start crond, stop services, restart services, or check if services are running.
+---
+
+# Devcontainer Services
+
+You operate inside the devcontainer. Do not use `docker exec`.
+
+## Start Backend (Python)
+
+```bash
+/services/start-backend.sh
+```
+
+Backend runs with debugpy on port 5678 for debugging. Takes ~5 seconds to be ready.
+
+## Start Frontend (nginx + PHP-FPM)
+
+```bash
+/services/start-php-fpm.sh &
+/services/start-nginx.sh &
+```
+
+Launches almost instantly.
+
+## Start Scheduler (CronD)
+
+```bash
+/services/start-crond.sh
+```
+
+## Stop All Services
+
+```bash
+pkill -f 'php-fpm83|nginx|crond|python3' || true
+```
+
+## Check Running Services
+
+```bash
+pgrep -a 'python3|nginx|php-fpm|crond'
+```
+
+## Service Ports
+
+- Frontend (nginx): 20211
+- Backend API: 20212
+- GraphQL: 20212
+- Debugpy: 5678

.github/skills/devcontainer-setup/SKILL.md

@@ -0,0 +1,36 @@
+---
+name: netalertx-idempotent—setup
+description: Reprovision and reset the devcontainer environment. Use this when asked to re-run startup, reprovision, setup devcontainer, fix permissions, or reset runtime state.
+---
+
+# Devcontainer Setup
+
+The setup script forcefully resets all runtime state. It is idempotent—every run wipes and recreates all relevant folders, symlinks, and files.
+
+## Command
+
+```bash
+/workspaces/NetAlertX/.devcontainer/scripts/setup.sh
+```
+
+## What It Does
+
+1. Kills all services (php-fpm, nginx, crond, python3)
+2. Mounts tmpfs ramdisks for `/tmp/log`, `/tmp/api`, `/tmp/run`, `/tmp/nginx`
+3. Creates critical subdirectories
+4. Links `/entrypoint.d` and `/app` symlinks
+5. Creates `/data`, `/data/config`, `/data/db` directories
+6. Creates all log files
+7. Runs `/entrypoint.sh` to start services
+8. Writes version to `.VERSION`
+
+## When to Use
+
+- After modifying setup scripts
+- After container rebuild
+- When environment is in broken state
+- After database reset
+
+## Philosophy
+
+No conditional logic. Everything is recreated unconditionally. If something doesn't work, run setup again.

.github/skills/docker-build/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: netalertx-docker-build
+description: Build Docker images for testing or production. Use this when asked to build container, build image, docker build, build test image, or launch production container.
+---
+
+# Docker Build
+
+## Build Unit Test Image
+
+Required after container/Dockerfile changes. Tests won't see changes until image is rebuilt.
+
+```bash
+docker buildx build -t netalertx-test .
+```
+
+Build time: ~30 seconds (or ~90s if venv stage changes)
+
+## Build and Launch Production Container
+
+Before launching, stop devcontainer services first to free ports.
+
+```bash
+cd /workspaces/NetAlertX
+docker compose up -d --build --force-recreate
+```
+
+## Pre-Launch Checklist
+
+1. Stop devcontainer services: `pkill -f 'php-fpm83|nginx|crond|python3'`
+2. Close VS Code forwarded ports
+3. Run the build command
+
+## Production Container Details
+
+- Image: `netalertx:latest`
+- Container name: `netalertx`
+- Network mode: host
+- Ports: 20211 (UI), 20212 (API/GraphQL)

.github/skills/docker-prune/SKILL.md

@@ -0,0 +1,32 @@
+---
+name: netalertx-docker-prune
+description: Clean up unused Docker resources. Use this when asked to prune docker, clean docker, remove unused images, free disk space, or docker cleanup. DANGEROUS operation. Requires human confirmation.
+---
+
+# Docker Prune
+
+**DANGER:** This destroys containers, images, volumes, and networks. Any stopped container will be wiped and data will be lost.
+
+## Command
+
+```bash
+/workspaces/NetAlertX/.devcontainer/scripts/confirm-docker-prune.sh
+```
+
+## What Gets Deleted
+
+- All stopped containers
+- All unused images
+- All unused volumes
+- All unused networks
+
+## When to Use
+
+- Disk space is low
+- Build cache is corrupted
+- Clean slate needed for testing
+- After many image rebuilds
+
+## Safety
+
+The script requires explicit confirmation before proceeding.

.github/skills/plugin-run-development/SKILL.md

@@ -0,0 +1,85 @@
+---
+name: netalertx-plugin-run-development
+description: Create and run NetAlertX plugins. Use this when asked to create plugin, run plugin, test plugin, plugin development, or execute plugin script.
+---
+
+# Plugin Development
+
+## Expected Workflow for Running Plugins
+
+1. Read this skill document for context and instructions.
+2. Find the plugin in `front/plugins/<code_name>/`.
+3. Read the plugin's `config.json` and `script.py` to understand its functionality and settings.
+4. Formulate and run the command: `python3 front/plugins/<code_name>/script.py`.
+5. Retrieve the result from the plugin log folder (`/tmp/log/plugins/last_result.<PREF>.log`) quickly, as the backend may delete it after processing.
+
+## Run a Plugin Manually
+
+```bash
+python3 front/plugins/<code_name>/script.py
+```
+
+Ensure `sys.path` includes `/app/front/plugins` and `/app/server` (as in the template).
+
+## Plugin Structure
+
+```
+front/plugins/<code_name>/
+├── config.json      # Manifest with settings
+├── script.py        # Main script
+└── ...
+```
+
+## Manifest Location
+
+`front/plugins/<code_name>/config.json`
+
+- `code_name` == folder name
+- `unique_prefix` drives settings and filenames (e.g., `ARPSCAN`)
+
+## Settings Pattern
+
+- `<PREF>_RUN`: execution phase
+- `<PREF>_RUN_SCHD`: cron-like schedule
+- `<PREF>_CMD`: script path
+- `<PREF>_RUN_TIMEOUT`: timeout in seconds
+- `<PREF>_WATCH`: columns to watch for changes
+
+## Data Contract
+
+Scripts write to `/tmp/log/plugins/last_result.<PREF>.log`
+
+**Important:** The backend will almost immediately process this result file and delete it after ingestion. If you need to inspect the output, run the plugin and immediately retrieve the result file before the backend processes it.
+
+Use `front/plugins/plugin_helper.py`:
+
+```python
+from plugin_helper import Plugin_Objects
+
+plugin_objects = Plugin_Objects()
+plugin_objects.add_object(...)  # During processing
+plugin_objects.write_result_file()  # Exactly once at end
+```
+
+## Execution Phases
+
+- `once`: runs once at startup
+- `schedule`: runs on cron schedule
+- `always_after_scan`: runs after every scan
+- `before_name_updates`: runs before name resolution
+- `on_new_device`: runs when new device detected
+- `on_notification`: runs when notification triggered
+
+## Plugin Formats
+
+| Format | Purpose | Runs |
+|--------|---------|------|
+| publisher | Send notifications | `on_notification` |
+| dev scanner | Create/manage devices | `schedule` |
+| name discovery | Discover device names | `before_name_updates` |
+| importer | Import from services | `schedule` |
+| system | Core functionality | `schedule` |
+
+## Starting Point
+
+Copy from `front/plugins/__template` and customize.

.github/skills/project-navigation/SKILL.md

@@ -0,0 +1,59 @@
+---
+name: about-netalertx-project-structure
+description: Navigate the NetAlertX codebase structure. Use this when asked about file locations, project structure, where to find code, or key paths.
+---
+
+# Project Navigation
+
+## Key Paths
+
+| Component | Path |
+|-----------|------|
+| Workspace root | `/workspaces/NetAlertX` |
+| Backend entry | `server/__main__.py` |
+| API server | `server/api_server/api_server_start.py` |
+| Plugin system | `server/plugin.py` |
+| Initialization | `server/initialise.py` |
+| Frontend | `front/` |
+| Frontend JS | `front/js/common.js` |
+| Frontend PHP | `front/php/server/*.php` |
+| Plugins | `front/plugins/` |
+| Plugin template | `front/plugins/__template` |
+| Database helpers | `server/db/db_helper.py` |
+| Device model | `server/models/device_instance.py` |
+| Messaging | `server/messaging/` |
+| Workflows | `server/workflows/` |
+
+## Architecture
+
+NetAlertX uses a frontend–backend architecture: the frontend runs on **PHP + Nginx** (see `front/`), the backend is implemented in **Python** (see `server/`), and scheduled tasks are managed by a **supercronic** scheduler that runs periodic jobs.
+
+## Runtime Paths
+
+| Data | Path |
+|------|------|
+| Config (runtime) | `/data/config/app.conf` |
+| Config (default) | `back/app.conf` |
+| Database | `/data/db/app.db` |
+| API JSON cache | `/tmp/api/*.json` |
+| Logs | `/tmp/log/` |
+| Plugin logs | `/tmp/log/plugins/` |
+
+## Environment Variables
+
+Use these instead of hardcoding paths:
+
+- `NETALERTX_DB`
+- `NETALERTX_LOG`
+- `NETALERTX_CONFIG`
+- `NETALERTX_DATA`
+- `NETALERTX_APP`
+
+## Documentation
+
+| Topic | Path |
+|-------|------|
+| Plugin development | `docs/PLUGINS_DEV.md` |
+| System settings | `docs/SETTINGS_SYSTEM.md` |
+| API docs | `docs/API_*.md` |
+| Debug guides | `docs/DEBUG_*.md` |

.github/skills/sample-data/SKILL.md

@@ -0,0 +1,31 @@
+---
+name: netalertx-sample-data
+description: Load synthetic device data into the devcontainer. Use this when asked to load sample devices, seed data, import test devices, populate database, or generate test data.
+---
+
+# Sample Data Loading
+
+Generates synthetic device inventory and imports it via the `/devices/import` API endpoint.
+
+## Command
+
+```bash
+cd /workspaces/NetAlertX/.devcontainer/scripts
+./load-devices.sh
+```
+
+## Environment
+
+- `CSV_PATH`: defaults to `/tmp/netalertx-devices.csv`
+
+## Prerequisites
+
+- Backend must be running
+- API must be accessible
+
+## What It Does
+
+1. Generates synthetic device records (MAC addresses, IPs, names, vendors)
+2. Creates CSV file at `$CSV_PATH`
+3. POSTs to `/devices/import` endpoint
+4. Devices appear in database and UI

.github/skills/settings-management/SKILL.md

@@ -0,0 +1,47 @@
+---
+name: netalertx-settings-management
+description: Manage NetAlertX configuration settings. Use this when asked to add setting, read config, get_setting_value, ccd, or configure options.
+---
+
+# Settings Management
+
+## Reading Settings
+
+```python
+from helper import get_setting_value
+
+value = get_setting_value('SETTING_NAME')
+```
+
+Never hardcode ports, secrets, or configuration values. Always use `get_setting_value()`.
+
+## Adding Core Settings
+
+Use `ccd()` in `server/initialise.py`:
+
+```python
+ccd('SETTING_NAME', 'default_value', 'description')
+```
+
+## Adding Plugin Settings
+
+Define in plugin's `config.json` manifest under the settings section.
+
+## Config Files
+
+| File | Purpose |
+|------|---------|
+| `/data/config/app.conf` | Runtime config (modified by app) |
+| `back/app.conf` | Default config (template) |
+
+## Environment Override
+
+Use `APP_CONF_OVERRIDE` environment variable for settings that must be set before startup.
+
+## Backend API URL
+
+For Codespaces, set `BACKEND_API_URL` to your Codespace URL:
+
+```
+BACKEND_API_URL=https://something-20212.app.github.dev/
+```

.github/skills/testing-workflow/SKILL.md

@@ -0,0 +1,55 @@
+---
+name: netalertx-testing-workflow
+description: Run and debug tests in the NetAlertX devcontainer. Use this when asked to run tests, check test failures, debug failing tests, or execute pytest.
+---
+
+# Testing Workflow
+
+## Pre-Flight Check (MANDATORY)
+
+Before running any tests, always check for existing failures first:
+
+1. Use the `testFailure` tool to gather current failure information
+2. Review the failures to understand what's already broken
+3. Only then proceed with test execution
+
+## Running Tests
+
+Use VS Code's testing interface or the `runTests` tool with appropriate parameters:
+
+- To run all tests: invoke runTests without file filter
+- To run specific test file: invoke runTests with the test file path
+- To run failed tests only: invoke runTests with `--lf` flag
+
+## Test Location
+
+Tests live in `test/` directory. App code is under `server/`.
+
+PYTHONPATH is preconfigured to include:
+- `/workspaces/NetAlertX`
+- `/opt/venv/lib/python3.12/site-packages`
+
+## Authentication in Tests
+
+Retrieve `API_TOKEN` using Python (not shell):
+
+```python
+from helper import get_setting_value
+token = get_setting_value("API_TOKEN")
+```
+
+## Troubleshooting 403 Forbidden
+
+1. Ensure backend is running (use devcontainer-services skill)
+2. Verify config loaded: `get_setting_value("API_TOKEN")` returns non-empty
+3. Re-run startup if needed (use devcontainer-setup skill)
+
+## Docker Test Image
+
+If container changes affect tests, rebuild the test image first:
+
+```bash
+docker buildx build -t netalertx-test .
+```
+
+This takes ~30 seconds unless venv stage changes (~90s).