Learning paths › Automation › Agents, async and model-driven workflows
Agents, async and model-driven workflows
Three strands in one module: model-driven workflows, where output decides the next step and a schema forces the answer's shape, with spans traced to see where the time went, two agents chained, and a Streamlit page over the top; NAPALM's drivers for a show command, a config backup and restore, normalised facts and a commit with a rollback timer; and Nornir coordinating inventory, host filters, labelled results and subtask order — ending as ncclient opens a NETCONF session and retrieves the running datastore. Explanation and checks; three written exercises.
Lessons
- Let model output decide the next step
- Await a coroutine so waits overlap
- Move a blocking API call off the event loop
- Trace spans to see where time went
- Chain two agents into a fixed pipeline
- Trade reasoning effort against latency and tokens
- Validate a network model at runtime
- Force the model to answer in a schema
- Wrap the agent in a Streamlit page
- Classify a config change as safe or dangerous
- Run a show command through a NAPALM driver
- Back up running config and restore it
- Commit with a timer that rolls back
- Read normalised facts with getters
- Nornir: coordinate inventory, tasks and results
- Initialise Nornir with or without config.yaml
- Read hosts, groups and defaults
- Target a subset of hosts with filters
- Label each block of task output
- Order subtasks inside one task function
- YANG models and the NETCONF that carries them
- Open a NETCONF session with ncclient
- Retrieve the running datastore
- Request one leaf with a subtree filter
More in Automation
- First scripts that configure devices
- NAPALM: structured data across vendors
- Concurrency, and the Python types device output returns
- Parsing show output and testing conditions
- Readable code, branching and loops
- Build the workstation and the toolchain
- Files, CSV, JSON and pickle
- Paramiko: SSH to devices from Python
- Netmiko at fleet scale
- Tuples, sets and dictionaries for device records
- Exceptions and your first classes
- Plan and build the automation lab
- Files, formatting and regular expressions
- Addresses, functions, modules and classes
- CSV inventories, logging and scheduling
- Netmiko, SCP and structured parsing
- NETCONF, YANG and RESTCONF
- NX-API, pyATS and Genie
- A small GenAI app, end to end
- macOS, Linux and the virtual lab
- Agile delivery and Python fundamentals
- REST APIs, authentication and requests
- Flask, Docker and Jenkins
- Managing a router with NETCONF and RESTCONF
- Ansible: inventory, ad-hoc commands and first playbooks
- Roles and zero-touch provisioning for Linux switches
- Installing the controller and applying a project
- Setting up Python 3
- The Ansible lab and the controller
- Vault, variables and config changes
- Parsing output and knowing which variable wins
- Jinja filters and PAN-OS automation
- Loops, callbacks, strategy and run control
- Roles, imports and writing your own module
- Running playbooks from AWX