# Automate Your Network

John Capobianco's field notes on networks, automation, and AI.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## My First Time on NPR: Are We Building Software We Can't Control?

DevFeed: [My First Time on NPR: Are We Building Software We Can't Control?](<https://devfeed.tech/articles/my-first-time-on-npr-are-we-building-software-we-can-t-control-133960.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/ai/wypr-thoughts-on-tech-and-things/>)

Author: John Capobianco

Published: 2026-11-13T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [Spec Driven Development](<https://devfeed.tech/topics/spec-driven-development.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [rate-limiting](<https://devfeed.tech/topics/rate-limiting.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [article](<https://devfeed.tech/tags/article.md>), [building](<https://devfeed.tech/tags/building.md>), [building-software](<https://devfeed.tech/tags/building-software.md>), [can](<https://devfeed.tech/tags/can.md>), [community](<https://devfeed.tech/tags/community.md>), [control](<https://devfeed.tech/tags/control.md>), [development](<https://devfeed.tech/tags/development.md>), [first](<https://devfeed.tech/tags/first.md>), [itential](<https://devfeed.tech/tags/itential.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [my](<https://devfeed.tech/tags/my.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [npr](<https://devfeed.tech/tags/npr.md>), [public](<https://devfeed.tech/tags/public.md>), [public-radio](<https://devfeed.tech/tags/public-radio.md>), [radio](<https://devfeed.tech/tags/radio.md>), [software](<https://devfeed.tech/tags/software.md>), [specification](<https://devfeed.tech/tags/specification.md>), [specification-driven-development](<https://devfeed.tech/tags/specification-driven-development.md>), [tech](<https://devfeed.tech/tags/tech.md>), [things](<https://devfeed.tech/tags/things.md>), [time](<https://devfeed.tech/tags/time.md>)

### AI overview

The interview's central takeaway is to use AI agents deliberately: begin with read access and triage, expand trust as earned, and keep humans involved in risky actions. The guest describes agents operating across tools and APIs, including a reported attempt to replicate NetClaw that they chose not to pursue further. They recommend specification-driven development, security, privacy, HR and finance input through an "AI council," and moving reliable agent work into deterministic workflows. The discussion also covers token costs and model choices, and cites the guest's on-air projection that agents could handle roughly 70% of network work by 2030, compared with about 1% today. The article summarizes a 49-minute interview and its examples; it does not establish the projection independently.

### Source excerpt

My national public radio debut on WYPR's Thoughts on Tech & Things with Jason Michael Perry: AI coding, specification-driven development, tokenomics, agents that escape their sandboxes, and a NetClaw check-in from my watch.

## Are We Building Software We Can't Control?

DevFeed: [Are We Building Software We Can't Control?](<https://devfeed.tech/articles/are-we-building-software-we-can-t-control-133961.md>)

Original publisher: [Read original article](<https://www.wypr.org/podcast/thoughts-on-tech-things/2026-11-13/are-we-building-software-we-cant-control>)

Author: Jason Michael Perry

Published: 2026-11-13T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Spec Driven Development](<https://devfeed.tech/topics/spec-driven-development.md>), [code security](<https://devfeed.tech/topics/code-security.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-security](<https://devfeed.tech/tags/agent-security.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [appearance](<https://devfeed.tech/tags/appearance.md>), [building](<https://devfeed.tech/tags/building.md>), [can](<https://devfeed.tech/tags/can.md>), [coding](<https://devfeed.tech/tags/coding.md>), [control](<https://devfeed.tech/tags/control.md>), [development](<https://devfeed.tech/tags/development.md>), [john](<https://devfeed.tech/tags/john.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [podcast](<https://devfeed.tech/tags/podcast.md>), [security](<https://devfeed.tech/tags/security.md>), [software](<https://devfeed.tech/tags/software.md>), [specification](<https://devfeed.tech/tags/specification.md>), [specification-driven-development](<https://devfeed.tech/tags/specification-driven-development.md>), [tech](<https://devfeed.tech/tags/tech.md>), [things](<https://devfeed.tech/tags/things.md>), [thoughts](<https://devfeed.tech/tags/thoughts.md>), [tokenomics](<https://devfeed.tech/tags/tokenomics.md>)

### AI overview

The episode examines how AI coding tools are changing software work and raising control and security concerns. Jason Michael Perry and John Capobianco discuss writing less code, specification-driven development, tokenomics, and engineers trusting code they have not reviewed. They also consider reports of agents circumventing restrictions, the risks of granting AI access to real infrastructure, and the challenge of controlling autonomous systems. The excerpt notes discussion of combining AI with traditional automation and whether humans may cease to be the best software writers; it does not provide detailed conclusions or specific security practices.

### Source excerpt

WYPR 88.1 FM - Podcast - John's national public radio debut on Thoughts on Tech & Things with Jason Michael Perry: AI coding, specification-driven development, tokenomics, agent security, and NetClaw on an Apple Watch.

## A Risk in the Clouds: NetClaw on GCP

DevFeed: [A Risk in the Clouds: NetClaw on GCP](<https://devfeed.tech/articles/a-risk-in-the-clouds-netclaw-on-gcp-131812.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/netclaw/netclaw-on-gcp/>)

Author: John Capobianco

Published: 2026-11-12T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [experience report](<https://devfeed.tech/topics/experience-report.md>), [Cloud Networking](<https://devfeed.tech/topics/cloud-networking.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Ollama](<https://devfeed.tech/topics/ollama.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [1-3](<https://devfeed.tech/tags/1-3.md>), [17](<https://devfeed.tech/tags/17.md>), [2](<https://devfeed.tech/tags/2.md>), [22](<https://devfeed.tech/tags/22.md>), [3](<https://devfeed.tech/tags/3.md>), [ai](<https://devfeed.tech/tags/ai.md>), [article](<https://devfeed.tech/tags/article.md>), [border](<https://devfeed.tech/tags/border.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [full](<https://devfeed.tech/tags/full.md>), [gcp](<https://devfeed.tech/tags/gcp.md>), [google-cloud](<https://devfeed.tech/tags/google-cloud.md>), [ha](<https://devfeed.tech/tags/ha.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [live](<https://devfeed.tech/tags/live.md>), [members](<https://devfeed.tech/tags/members.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [one](<https://devfeed.tech/tags/one.md>), [openshell](<https://devfeed.tech/tags/openshell.md>), [region](<https://devfeed.tech/tags/region.md>), [risk](<https://devfeed.tech/tags/risk.md>), [teardown](<https://devfeed.tech/tags/teardown.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

The article reports that NetClaw Borders and Members ran in GCP, on-premises, hybrid setups, and a live three-region Risk, but warns the system is not production-ready because Border URLs have no authentication: anyone with one can use the HUD and Chat and spend the model API key. In the tested design, Members dial out to Borders; multiple Borders share one execution authority, and loss of quorum blocks new protected work rather than authorizing duplicate execution. Multi-region tests exposed latency-sensitive admission windows, slow failover, and a distant two-region quorum that could not admit work. OpenShell with DefenseClaw added audited blocking and sandbox isolation, with measured overhead below half a percent of the 4,500 ms admission window, though one leadership timing issue remained unexplained. The tested workloads were read-only. The article recommends experimenting in a throwaway GCP project, setting budget alerts (which do not cap spending), keeping Border URLs private, and fully deleting resources afterward. Users, SSO, and RBAC are roadmap work, not current features.

### Source excerpt

From one Border to a three-region Risk: the full NetClaw on GCP story, R01-R09C, with cloud and hybrid Members, HA, OpenShell, DefenseClaw, live evidence and lessons from teardown.

## The Engineer Becomes the Agent Manager: My Techstrong TV Interview

DevFeed: [The Engineer Becomes the Agent Manager: My Techstrong TV Interview](<https://devfeed.tech/articles/the-engineer-becomes-the-agent-manager-my-techstrong-tv-interview-119698.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/ai/techstrong-tv-agent-manager/>)

Author: John Capobianco

Published: 2026-11-05T12:00:00Z

Content type: opinion

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [engineering-leadership](<https://devfeed.tech/topics/engineering-leadership.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>)

Tags: [acceptance](<https://devfeed.tech/tags/acceptance.md>), [accountability](<https://devfeed.tech/tags/accountability.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [alan-shimel](<https://devfeed.tech/tags/alan-shimel.md>), [article](<https://devfeed.tech/tags/article.md>), [assessment](<https://devfeed.tech/tags/assessment.md>), [audit](<https://devfeed.tech/tags/audit.md>), [automation](<https://devfeed.tech/tags/automation.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [community](<https://devfeed.tech/tags/community.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [engineers](<https://devfeed.tech/tags/engineers.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [interview](<https://devfeed.tech/tags/interview.md>), [itential](<https://devfeed.tech/tags/itential.md>), [manager](<https://devfeed.tech/tags/manager.md>), [managing-ai-agents](<https://devfeed.tech/tags/managing-ai-agents.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [techstrong-tv](<https://devfeed.tech/tags/techstrong-tv.md>), [the-result](<https://devfeed.tech/tags/the-result.md>)

### AI overview

The article argues that engineers managing AI agents should define each agent's responsibility, authority, escalation conditions, user, and owner, then design explicit handoffs so downstream participants can inspect the evidence. It recommends enforcing permissions and approvals in the execution system, testing prohibited actions and stale approvals, and evaluating reviewer agents on both flawed and sound proposals. One cited Jev and FlowAgents demonstration included a rollback request and incomplete recovery requiring operator attention, alongside successful lab runs, including one with human approval. To assess cost, compare model use, tool calls, elapsed time, and human review against accepted results, including retries and corrections. After launch, track ownership, access, consumers, evaluations, and review dates; plan for escalation workload and retire access when agents are removed. The article presents these as practical design recommendations and examples, not a universal architecture.

### Source excerpt

My Techstrong TV debut with Alan Shimel, and what managing AI agents demands of engineers: clear responsibilities, meaningful evaluation, and accountability for the result.

## Connect NetClaw Mobile from the HUD

DevFeed: [Connect NetClaw Mobile from the HUD](<https://devfeed.tech/articles/connect-netclaw-mobile-from-the-hud-119699.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/netclaw/mobile-qr-enrollment/>)

Author: John Capobianco

Published: 2026-11-05T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [mobile-app-development](<https://devfeed.tech/topics/mobile-app-development.md>), [Mobile Testing](<https://devfeed.tech/topics/mobile-testing.md>), [Visual Studio Code](<https://devfeed.tech/topics/visual-studio-code.md>), [VS Code Extension](<https://devfeed.tech/topics/vscode-extension.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [article](<https://devfeed.tech/tags/article.md>), [border](<https://devfeed.tech/tags/border.md>), [button](<https://devfeed.tech/tags/button.md>), [connect](<https://devfeed.tech/tags/connect.md>), [device](<https://devfeed.tech/tags/device.md>), [device-identity](<https://devfeed.tech/tags/device-identity.md>), [iphone](<https://devfeed.tech/tags/iphone.md>), [mobile](<https://devfeed.tech/tags/mobile.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [qr-code](<https://devfeed.tech/tags/qr-code.md>), [scan](<https://devfeed.tech/tags/scan.md>), [vs-code](<https://devfeed.tech/tags/vs-code.md>)

### AI overview

NetClaw 1.18.6 and VS Code extension 1.6.1 add "Generate QR code" controls to their Mobile panels, making phone enrollment available from the active workspace. The user selects a Border, generates a code, then scans it in NetClaw Mobile; each code enrolls one device and expires after two minutes. If the response is lost, check the original enrollment status before retrying. The Border's listener and trusted transport must be ready. After enrollment, the app reconnects using its saved device identity, while existing key checks, destination binding, and operation review requirements remain in place. The target phone changes require NetClaw Mobile 1.0.4 build 7, which was awaiting Apple review; the listing then offered 1.0.2. Acceptance runs used the HUD, extension, and iPhone simulator, but do not establish physical camera or hardware Secure Enclave behavior.

### Source excerpt

Generate an enrollment QR code in the NetClaw HUD or VS Code Mobile panel, scan it with your phone, and reconnect with the same saved device identity.

## IT Is Becoming an HR Department for AI Agents

DevFeed: [IT Is Becoming an HR Department for AI Agents](<https://devfeed.tech/articles/it-is-becoming-an-hr-department-for-ai-agents-119700.md>)

Original publisher: [Read original article](<https://techstrong.tv/videos/interviews/it-hr-department-ai-agents>)

Author: Alan Shimel

Published: 2026-11-03T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>)

Tags: [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-governance](<https://devfeed.tech/tags/ai-governance.md>), [alan-shimel](<https://devfeed.tech/tags/alan-shimel.md>), [appearance](<https://devfeed.tech/tags/appearance.md>), [automation](<https://devfeed.tech/tags/automation.md>), [decision-making-models](<https://devfeed.tech/tags/decision-making-models.md>), [devrel](<https://devfeed.tech/tags/devrel.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [governance](<https://devfeed.tech/tags/governance.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [infrastructure-automation](<https://devfeed.tech/tags/infrastructure-automation.md>), [interview](<https://devfeed.tech/tags/interview.md>), [interviews](<https://devfeed.tech/tags/interviews.md>), [it](<https://devfeed.tech/tags/it.md>), [itential](<https://devfeed.tech/tags/itential.md>), [john-capobianco](<https://devfeed.tech/tags/john-capobianco.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [model-context-protocol](<https://devfeed.tech/tags/model-context-protocol.md>), [multi-agent-orchestration](<https://devfeed.tech/tags/multi-agent-orchestration.md>), [netdevops](<https://devfeed.tech/tags/netdevops.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [techstrong-tv](<https://devfeed.tech/tags/techstrong-tv.md>)

### AI overview

IT infrastructure teams can manage AI agents as a coordinated workforce: an orchestrator delegates planning, review, configuration, testing, and communication to agents with models and tools suited to each task. In Capobianco's example, a decision model scores a change plan and flags gaps, a human approves it, and a narrowly scoped configuration agent applies it; a low cost open weight model handles messages. Itential says its platform adds role based access control, audit, and authentication to move agents toward production use. The interview argues that specialized decision models can check evidence, proposed actions, and results before agents proceed, supporting more reliable decisions and guardrails as autonomy increases.

### Source excerpt

Techstrong TV - Interview - John joins Alan Shimel for his Techstrong TV debut to discuss his engineering journey, managing AI agents, and governance for infrastructure operations.

## Your AI Agent Needs an Operating Environment. Who's Engineering It?

DevFeed: [Your AI Agent Needs an Operating Environment. Who's Engineering It?](<https://devfeed.tech/articles/your-ai-agent-needs-an-operating-environment-who-s-engineering-it-117836.md>)

Original publisher: [Read original article](<https://devops.com/your-ai-agent-needs-an-operating-environment-whos-engineering-it>)

Author: John Capobianco

Published: 2026-11-03T12:00:00Z

Content type: opinion

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Access Control](<https://devfeed.tech/topics/access-control.md>), [knowledge-management](<https://devfeed.tech/topics/knowledge-management.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-harness](<https://devfeed.tech/tags/agent-harness.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [allowed](<https://devfeed.tech/tags/allowed.md>), [application](<https://devfeed.tech/tags/application.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [article](<https://devfeed.tech/tags/article.md>), [completion](<https://devfeed.tech/tags/completion.md>), [context](<https://devfeed.tech/tags/context.md>), [devops](<https://devfeed.tech/tags/devops.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [engineers](<https://devfeed.tech/tags/engineers.md>), [environment](<https://devfeed.tech/tags/environment.md>), [harness](<https://devfeed.tech/tags/harness.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [model](<https://devfeed.tech/tags/model.md>), [operating-environment](<https://devfeed.tech/tags/operating-environment.md>), [techstrong-council](<https://devfeed.tech/tags/techstrong-council.md>), [test](<https://devfeed.tech/tags/test.md>), [tools](<https://devfeed.tech/tags/tools.md>)

### AI overview

An infrastructure agent's safety and reliability depend on the operating environment engineers provide, not just the model. The article recommends starting with a bounded task, scoped diagnostic tools and validated change capabilities, task-specific context with clear freshness requirements, and controls that enforce approvals and verification. Procedures can guide an agent, but execution systems must block changes when mandatory requirements are unmet and define what happens when verification fails. Teams should set evidence-based completion criteria, test rejection, timeout, partial-completion, and recovery scenarios, and ensure reports match execution records. They can build, adopt, or combine platforms; the relevant responsibilities include integration, authorization, testing, observability, and maintenance. Test the agent and its environment together as models, tools, and procedures change.

### Source excerpt

DevOps.com - Article - John explains why infrastructure agents need scoped tools, current context, enforced approvals, and evidence of completion, and why engineers must own and test the operating environment around the model.

## Building With Lumen: Engineers, FlowAgents, and a Summit Full of Ideas

DevFeed: [Building With Lumen: Engineers, FlowAgents, and a Summit Full of Ideas](<https://devfeed.tech/articles/building-with-lumen-engineers-flowagents-and-a-summit-full-of-ideas-110093.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/itential/lumen-innovation-summit-2026/>)

Author: John Capobianco

Published: 2026-10-30T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [developer-relations](<https://devfeed.tech/topics/developer-relations.md>), [smolagents](<https://devfeed.tech/topics/smolagents.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [archive](<https://devfeed.tech/tags/archive.md>), [article](<https://devfeed.tech/tags/article.md>), [audit](<https://devfeed.tech/tags/audit.md>), [audit-trail](<https://devfeed.tech/tags/audit-trail.md>), [blog](<https://devfeed.tech/tags/blog.md>), [community](<https://devfeed.tech/tags/community.md>), [flowagents](<https://devfeed.tech/tags/flowagents.md>), [flowai](<https://devfeed.tech/tags/flowai.md>), [innovation](<https://devfeed.tech/tags/innovation.md>), [itential](<https://devfeed.tech/tags/itential.md>), [lumen](<https://devfeed.tech/tags/lumen.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [october](<https://devfeed.tech/tags/october.md>), [photos](<https://devfeed.tech/tags/photos.md>), [recap](<https://devfeed.tech/tags/recap.md>), [reflections](<https://devfeed.tech/tags/reflections.md>), [summit](<https://devfeed.tech/tags/summit.md>), [trust](<https://devfeed.tech/tags/trust.md>), [workshops](<https://devfeed.tech/tags/workshops.md>)

### AI overview

At Lumen's Innovation Summit, hands-on workshops put FlowAgents to work on engineering tasks: in a 45-minute session, 31 engineers ran 40 agents across 149 sessions, with 127 completions, 19 failures, and three cancellations. The recap stresses that completion alone does not show whether a session achieved its intended result. Across two workshop days, participants progressed from individual agents to examining existing workflows and coordinating agents around an approved change, including reverse-engineering seven inherited workflows into specifications. Greg Freeman's account highlights the broader summit's 200-plus engineers, 45 breakouts, and live demonstrations. Marc Montez frames the operating model as "AI decides, workflows execute," pairing agent interpretation and investigation with predictable execution, approvals, and audit trails. The excerpt describes the event and workshops but gives no complete date range.

### Source excerpt

Photos and reflections from Lumen's October Innovation Summit: hands-on FlowAI workshops, Greg Freeman's recap, Marc Montez on trust, and my full Itential write-up.

## NetClaw on Heavy Networking: An Open Source Network Assistant

DevFeed: [NetClaw on Heavy Networking: An Open Source Network Assistant](<https://devfeed.tech/articles/netclaw-on-heavy-networking-an-open-source-network-assistant-110096.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/netclaw/packet-pushers-heavy-networking/>)

Author: John Capobianco

Published: 2026-10-30T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [AI Networking](<https://devfeed.tech/topics/ai-networking.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [google-adk](<https://devfeed.tech/topics/google-adk.md>)

Tags: [actions](<https://devfeed.tech/tags/actions.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [article](<https://devfeed.tech/tags/article.md>), [cisco](<https://devfeed.tech/tags/cisco.md>), [community](<https://devfeed.tech/tags/community.md>), [draft](<https://devfeed.tech/tags/draft.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [environment](<https://devfeed.tech/tags/environment.md>), [experiment](<https://devfeed.tech/tags/experiment.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [governance](<https://devfeed.tech/tags/governance.md>), [lab](<https://devfeed.tech/tags/lab.md>), [layers](<https://devfeed.tech/tags/layers.md>), [maintainers](<https://devfeed.tech/tags/maintainers.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [network](<https://devfeed.tech/tags/network.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [networking](<https://devfeed.tech/tags/networking.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [protection](<https://devfeed.tech/tags/protection.md>)

### AI overview

NetClaw's excerpted discussion emphasizes trust through explicit approval: actions beyond reading network state require approval, with operational rules informed by engineering mistakes and outages. It describes Cisco DefenseClaw and NVIDIA OpenShell as additional protections to configure and evaluate in each environment, and recommends starting in a lab with read-only tasks while learning the tools' boundaries. NetClaw federation, called NCFED, is an Internet-Draft open for development and review. The project also keeps user stories, requirements, plans, and tests alongside implementation to help contributors understand the work. Maintainers, contributors, and engineers testing it in their labs are shaping the project.

### Source excerpt

Back with Ethan Banks and Drew Conry-Murray on Packet Pushers to talk NetClaw: MCP, explicit approvals, layered protection, federation, and the community building it.

## NetClaw returns to Packet Pushers

DevFeed: [NetClaw returns to Packet Pushers](<https://devfeed.tech/articles/netclaw-returns-to-packet-pushers-110106.md>)

Original publisher: [Read original article](<https://www.linkedin.com/posts/john-capobianco-644a1515_netclaw-networkautomation-agenticai-activity-7521962635014283265-GaUw>)

Author: John Capobianco

Published: 2026-10-30T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [AI Networking](<https://devfeed.tech/topics/ai-networking.md>), [Automation](<https://devfeed.tech/topics/automation.md>), [Network Operations](<https://devfeed.tech/topics/network-operations.md>), [Fair Source](<https://devfeed.tech/topics/fair-source.md>), [Sports](<https://devfeed.tech/topics/sports.md>), [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>)

Tags: [2019](<https://devfeed.tech/tags/2019.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [air](<https://devfeed.tech/tags/air.md>), [air-gapped](<https://devfeed.tech/tags/air-gapped.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automate](<https://devfeed.tech/tags/automate.md>), [automation](<https://devfeed.tech/tags/automation.md>), [banks](<https://devfeed.tech/tags/banks.md>), [community](<https://devfeed.tech/tags/community.md>), [contributors](<https://devfeed.tech/tags/contributors.md>), [john](<https://devfeed.tech/tags/john.md>), [john-capobianco](<https://devfeed.tech/tags/john-capobianco.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [network](<https://devfeed.tech/tags/network.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [networking](<https://devfeed.tech/tags/networking.md>), [packet](<https://devfeed.tech/tags/packet.md>), [social](<https://devfeed.tech/tags/social.md>)

### AI overview

The author says NetClaw makes network automation easier to try: instead of months of book study and lab work, users can add eight lines of JSON to an agent. The project has three co-maintainers, 16+ contributors, more than 200 documented specs, and integrations being built and tested by named network vendors. Its safety approach includes a "soul file" of operational rules and explicit approval for production actions beyond reads; the author also cites Cisco DefenseClaw and NVIDIA OpenShell as additional protection. NetClaw supports frontier or air-gapped open-weight models, including on Raspberry Pi, and offers a VS Code extension plus mobile and watch access. The author recommends starting read-only in a lab. These are claims from the supplied excerpt.

### Source excerpt

John Capobianco - LinkedIn - John reflects on NetClaw's community, its contributors, and making useful network automation easier to try alongside the new Heavy Networking episode.

## HN848: NetClaw: Your Open Source Network Assistant

DevFeed: [HN848: NetClaw: Your Open Source Network Assistant](<https://devfeed.tech/articles/hn848-netclaw-your-open-source-network-assistant-110098.md>)

Original publisher: [Read original article](<https://packetpushers.net/podcasts/heavy-networking/hn848-netclaw-your-open-source-network-assistant/>)

Author: Heavy Networking Ep ; October

Published: 2026-10-30T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Open Source](<https://devfeed.tech/topics/open-source.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [appearance](<https://devfeed.tech/tags/appearance.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automation](<https://devfeed.tech/tags/automation.md>), [flowai](<https://devfeed.tech/tags/flowai.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [network](<https://devfeed.tech/tags/network.md>), [networking](<https://devfeed.tech/tags/networking.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [podcast](<https://devfeed.tech/tags/podcast.md>), [production](<https://devfeed.tech/tags/production.md>), [security](<https://devfeed.tech/tags/security.md>), [security-architecture](<https://devfeed.tech/tags/security-architecture.md>)

### AI overview

NetClaw is John Capobianco's open source automation assistant for network engineers, built around MCP with a security architecture intended to protect production systems. In this Heavy Networking episode, Capobianco explains why and how he built it and discusses potential day-to-day uses. The supplied excerpt does not describe its operating rules, protections, or specific workflows.

### Source excerpt

Packet Pushers - Podcast - John joins Ethan Banks and Drew Conry-Murray on Heavy Networking to discuss NetClaw, operating rules, layered protection, federation, and community development.

## From a Converted Configuration to a Migration You Can Test

DevFeed: [From a Converted Configuration to a Migration You Can Test](<https://devfeed.tech/articles/from-a-converted-configuration-to-a-migration-you-can-test-110095.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/netclaw/netconverter-migration-rehearsals/>)

Author: John Capobianco

Published: 2026-10-30T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Test automation](<https://devfeed.tech/topics/test-automation.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [article](<https://devfeed.tech/tags/article.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [change](<https://devfeed.tech/tags/change.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [conversion](<https://devfeed.tech/tags/conversion.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [migration](<https://devfeed.tech/tags/migration.md>), [netclaw](<https://devfeed.tech/tags/netclaw.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [target](<https://devfeed.tech/tags/target.md>), [test](<https://devfeed.tech/tags/test.md>), [validation](<https://devfeed.tech/tags/validation.md>), [virtual](<https://devfeed.tech/tags/virtual.md>)

### AI overview

NetClaw is being integrated with NetConverter to carry network migrations from conversion through virtual-target rehearsal and change evidence, while requiring engineer review at key handoffs. Before submission, NetClaw freezes the source file and shows its hash, size, destination, account, operation, direction, and options for approval; changes invalidate approval, and a lost response is treated as an unknown outcome to prevent an automatic duplicate job. Candidate files can remain private, and applying one to a designated lab endpoint requires baseline and rollback preparation plus a short-lived, one-use capability. Controlled tests cover MCP transports, approval, artifacts, application, failures, and lab orchestration, but the CML peer is a fixture. Live NetConverter account conversion and Junos acceptance remain unqualified: account enablement, image and management-path checks, and live commit/readback are pending. EIGRP is not translated to Junos, so that migration requires explicit protocol redesign. The planned IOS-XE-to-Junos OSPF demonstration has not yet run.

### Source excerpt

How NetConverter and NetClaw can connect reviewed conversion artifacts with virtual-target rehearsals, validation and change evidence. Live conversion qualification is pending.

## The Testbed Becomes Part of the Conversation: Dynamic Inventory in pyATS MCP

DevFeed: [The Testbed Becomes Part of the Conversation: Dynamic Inventory in pyATS MCP](<https://devfeed.tech/articles/the-testbed-becomes-part-of-the-conversation-dynamic-inventory-in-pyats-mcp-110097.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/pyats/dynamic-testbeds-mcp/>)

Author: John Capobianco

Published: 2026-10-30T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [internal developer platform automation](<https://devfeed.tech/topics/internal-developer-platform-automation.md>)

Tags: [activation](<https://devfeed.tech/tags/activation.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [article](<https://devfeed.tech/tags/article.md>), [automation](<https://devfeed.tech/tags/automation.md>), [builder](<https://devfeed.tech/tags/builder.md>), [catalyst-center](<https://devfeed.tech/tags/catalyst-center.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [credentials](<https://devfeed.tech/tags/credentials.md>), [csv](<https://devfeed.tech/tags/csv.md>), [declarative](<https://devfeed.tech/tags/declarative.md>), [filesystem](<https://devfeed.tech/tags/filesystem.md>), [fleet-operations](<https://devfeed.tech/tags/fleet-operations.md>), [genie](<https://devfeed.tech/tags/genie.md>), [http](<https://devfeed.tech/tags/http.md>), [inventory](<https://devfeed.tech/tags/inventory.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [memory](<https://devfeed.tech/tags/memory.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [pyats](<https://devfeed.tech/tags/pyats.md>), [testbed](<https://devfeed.tech/tags/testbed.md>), [tools](<https://devfeed.tech/tags/tools.md>)

### AI overview

pyATS MCP can build a testbed in server memory from inventory systems or structured inputs, let an operator inspect and activate it, then use it with existing pyATS tools. At the cited revision, it exposes 34 tools, including eight dynamic-testbed tools, with native readers for Itential Inventory Manager, NetBox, Nautobot and Catalyst Center, plus generic REST and CSV, JSON, YAML or XML inputs with field mapping. A valid testbed confirms acceptable structure, not device reachability, working credentials or successful commands. Credential and access boundaries still need configuration and review. The dynamic registry is process-local and is lost on server restart; some subprocess workflows create private temporary testbed files.

### Source excerpt

Part one: pyATS MCP can build an in-memory testbed from inventory systems and structured data, then make it available to its existing network automation tools.

## Learning in Public: Elliot Conner in the VibeOps Lounge

DevFeed: [Learning in Public: Elliot Conner in the VibeOps Lounge](<https://devfeed.tech/articles/learning-in-public-elliot-conner-in-the-vibeops-lounge-110092.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/ai/vibeops-lounge-elliot-conner/>)

Author: John Capobianco

Published: 2026-10-29T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Learning](<https://devfeed.tech/topics/learning.md>), [AI Networking](<https://devfeed.tech/topics/ai-networking.md>), [AI in IT operations](<https://devfeed.tech/topics/ai-in-it-operations.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [access-controls](<https://devfeed.tech/tags/access-controls.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [also](<https://devfeed.tech/tags/also.md>), [article](<https://devfeed.tech/tags/article.md>), [at-t](<https://devfeed.tech/tags/at-t.md>), [automation](<https://devfeed.tech/tags/automation.md>), [automation-tools](<https://devfeed.tech/tags/automation-tools.md>), [build](<https://devfeed.tech/tags/build.md>), [building](<https://devfeed.tech/tags/building.md>), [building-automation](<https://devfeed.tech/tags/building-automation.md>), [can](<https://devfeed.tech/tags/can.md>), [community](<https://devfeed.tech/tags/community.md>), [game](<https://devfeed.tech/tags/game.md>), [learning](<https://devfeed.tech/tags/learning.md>), [network](<https://devfeed.tech/tags/network.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [operations](<https://devfeed.tech/tags/operations.md>), [planner](<https://devfeed.tech/tags/planner.md>), [public](<https://devfeed.tech/tags/public.md>), [troubleshooting](<https://devfeed.tech/tags/troubleshooting.md>), [vibeops](<https://devfeed.tech/tags/vibeops.md>)

### AI overview

Elliot Conner's three demos show practical ways to learn network automation and AI by building around familiar problems: a Containerlab operations interface covering access controls, NetBox, change management, rollback, and lab inspection; Net Planner, an AI-assisted tool for planning with cost comparisons, vendor quotes, and reports; and an interactive CLI troubleshooting game with VLAN and OSPF problems. The conversation traces his move from CLI experience into Python, pyATS, and MCP experiments, and emphasizes building a public GitHub portfolio and sharing what you learn. The excerpt describes the demos but does not establish their implementation details or results.

### Source excerpt

VibeOps Lounge 007: Elliot Conner brings a Containerlab operations dashboard, an AI network planner, and a troubleshooting game to a conversation about learning by building.

## VibeOps Lounge 007 - Learning Network Automation & AI in Public with Elliot Conner

DevFeed: [VibeOps Lounge 007 - Learning Network Automation & AI in Public with Elliot Conner](<https://devfeed.tech/articles/vibeops-lounge-007-learning-network-automation-ai-in-public-with-elliot-conner-110113.md>)

Original publisher: [Read original article](<https://www.youtube.com/watch?v=WnXN88AJD8k>)

Published: 2026-10-29T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Network Operations](<https://devfeed.tech/topics/network-operations.md>), [Learning](<https://devfeed.tech/topics/learning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [camera-phone](<https://devfeed.tech/tags/camera-phone.md>), [dashboard](<https://devfeed.tech/tags/dashboard.md>), [demo](<https://devfeed.tech/tags/demo.md>), [discuss](<https://devfeed.tech/tags/discuss.md>), [free](<https://devfeed.tech/tags/free.md>), [game](<https://devfeed.tech/tags/game.md>), [interactive](<https://devfeed.tech/tags/interactive.md>), [john-capobianco](<https://devfeed.tech/tags/john-capobianco.md>), [learning](<https://devfeed.tech/tags/learning.md>), [network](<https://devfeed.tech/tags/network.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [network-troubleshooting](<https://devfeed.tech/tags/network-troubleshooting.md>), [operations](<https://devfeed.tech/tags/operations.md>), [planner](<https://devfeed.tech/tags/planner.md>), [public](<https://devfeed.tech/tags/public.md>), [sharing](<https://devfeed.tech/tags/sharing.md>), [troubleshooting](<https://devfeed.tech/tags/troubleshooting.md>), [upload](<https://devfeed.tech/tags/upload.md>), [video](<https://devfeed.tech/tags/video.md>), [video-phone](<https://devfeed.tech/tags/video-phone.md>)

### AI overview

Elliot Conner joins John Capobianco to discuss learning in public and demonstrate three network automation projects: a Containerlab operations dashboard, Net Planner, and an interactive network troubleshooting game.

### Source excerpt

John Capobianco - Lounge episode - Elliot Conner joins John to discuss learning in public and demo a Containerlab operations dashboard, Net Planner, and an interactive network troubleshooting game.

## Inside Lumen's Innovation Summit: 40 Agents in 45 Minutes With Itential FlowAI

DevFeed: [Inside Lumen's Innovation Summit: 40 Agents in 45 Minutes With Itential FlowAI](<https://devfeed.tech/articles/inside-lumen-s-innovation-summit-40-agents-in-45-minutes-with-itential-flowai-110101.md>)

Original publisher: [Read original article](<https://www.itential.com/resource/blog/lumen-innovation-summit-forty-agents-flowai/>)

Author: John Capobianco

Published: 2026-10-26T12:00:00Z

Content type: news

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [AI for customer service automation](<https://devfeed.tech/topics/ai-for-customer-service-automation.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>), [Multi-step AI processes](<https://devfeed.tech/topics/multi-step-ai-processes.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [agent-orchestration](<https://devfeed.tech/tags/agent-orchestration.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-reasoning](<https://devfeed.tech/tags/ai-reasoning.md>), [analyzer](<https://devfeed.tech/tags/analyzer.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [article](<https://devfeed.tech/tags/article.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [automation](<https://devfeed.tech/tags/automation.md>), [autopilot](<https://devfeed.tech/tags/autopilot.md>), [documentation](<https://devfeed.tech/tags/documentation.md>), [flowai](<https://devfeed.tech/tags/flowai.md>), [innovation](<https://devfeed.tech/tags/innovation.md>), [itential](<https://devfeed.tech/tags/itential.md>), [itential-flowai](<https://devfeed.tech/tags/itential-flowai.md>), [lumen](<https://devfeed.tech/tags/lumen.md>), [minutes](<https://devfeed.tech/tags/minutes.md>), [summit](<https://devfeed.tech/tags/summit.md>), [workflow](<https://devfeed.tech/tags/workflow.md>), [workshops](<https://devfeed.tech/tags/workshops.md>)

### AI overview

At Lumen's Innovation Summit, 31 engineers built or ran FlowAI agents, generating 149 sessions across 40 agents in the busiest 45 minutes; 127 completed, 19 failed, and 3 were canceled. The workshop used read-only network and inventory examples, while change demonstrations included checks, rollback, and human review. A second workshop used Itential Builder Skills and Spec-Driven Development to create workflows and turn seven inherited workflows into readable Markdown specifications. A third demonstrated a six-agent team for loopback provisioning, with defined roles, evidence checks, decision gates, and Work Center human approval. The article connects these experiments to Lumen's existing deterministic workflows: agents can interpret requests and coordinate work, while governed workflows, permissions, verification, and approvals handle infrastructure changes. It also distinguishes reasoning for uncertain tasks from deterministic workflows for repeatable ones, with a path to formalize recurring agent work as specifications and reusable assets.

### Source excerpt

Itential - Article - John's field report from Lumen's hands-on FlowAI workshops, including session outcomes, inherited workflow documentation, and human-approved agent coordination.

## John's Lumen reflection and the summit video memento

DevFeed: [John's Lumen reflection and the summit video memento](<https://devfeed.tech/articles/john-s-lumen-reflection-and-the-summit-video-memento-110105.md>)

Original publisher: [Read original article](<https://www.linkedin.com/posts/john-capobianco-644a1515_flowai-networkautomation-agenticops-activity-7520564866747678720-jRyl>)

Author: Marc Montez

Published: 2026-10-26T12:00:00Z

Content type: opinion

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [Claude](<https://devfeed.tech/topics/claude.md>)

Tags: [2025](<https://devfeed.tech/tags/2025.md>), [3](<https://devfeed.tech/tags/3.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [analyzer](<https://devfeed.tech/tags/analyzer.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [assessment](<https://devfeed.tech/tags/assessment.md>), [john-capobianco](<https://devfeed.tech/tags/john-capobianco.md>), [linkedin](<https://devfeed.tech/tags/linkedin.md>), [lumen](<https://devfeed.tech/tags/lumen.md>), [post](<https://devfeed.tech/tags/post.md>), [question](<https://devfeed.tech/tags/question.md>), [shares](<https://devfeed.tech/tags/shares.md>), [social](<https://devfeed.tech/tags/social.md>), [summit](<https://devfeed.tech/tags/summit.md>), [team](<https://devfeed.tech/tags/team.md>), [thanks](<https://devfeed.tech/tags/thanks.md>), [video](<https://devfeed.tech/tags/video.md>), [workflow](<https://devfeed.tech/tags/workflow.md>), [workshop](<https://devfeed.tech/tags/workshop.md>)

### AI overview

At Lumen's Innovation Summit, an engineer who inherited seven workflows asked whether they could be understood in reverse. The post says Itential's workflow-to-spec skill points Claude Code at an existing workflow and produces a readable Markdown spec. In a 45-minute workshop, 31 engineers built 40 FlowAgents, focused on interface and device status, inventory and topology, config changes, and drift and source-of-truth audits. One participant chained three agents into a router assessment. Of 127 sessions, 19 failed and 3 were canceled, underscoring the post's call to pair speed with checks, approvals, and rollback. Lumen's stated principle is "AI decides, workflows execute," relying on its governed workflows as a foundation for agent actions.

### Source excerpt

John Capobianco - LinkedIn - John shares the inherited-workflow question, the workshop outcomes, and thanks the Lumen team. The original post includes John Armstrong's video memento.

## Itential: engineers put FlowAI to work at Lumen

DevFeed: [Itential: engineers put FlowAI to work at Lumen](<https://devfeed.tech/articles/itential-engineers-put-flowai-to-work-at-lumen-110104.md>)

Original publisher: [Read original article](<https://www.linkedin.com/posts/itential_what-happens-when-network-engineers-get-hands-on-activity-7520557129980620800-9gxm>)

Author: Itential

Published: 2026-10-26T12:00:00Z

Content type: news

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [Spec Driven Development](<https://devfeed.tech/topics/spec-driven-development.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agents](<https://devfeed.tech/tags/agents.md>), [changes](<https://devfeed.tech/tags/changes.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [deterministic-workflows](<https://devfeed.tech/tags/deterministic-workflows.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [execution](<https://devfeed.tech/tags/execution.md>), [flowagents](<https://devfeed.tech/tags/flowagents.md>), [flowai](<https://devfeed.tech/tags/flowai.md>), [innovation](<https://devfeed.tech/tags/innovation.md>), [inventory](<https://devfeed.tech/tags/inventory.md>), [itential](<https://devfeed.tech/tags/itential.md>), [itential-flowai](<https://devfeed.tech/tags/itential-flowai.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [link](<https://devfeed.tech/tags/link.md>), [lumen](<https://devfeed.tech/tags/lumen.md>), [network](<https://devfeed.tech/tags/network.md>), [partners](<https://devfeed.tech/tags/partners.md>), [results](<https://devfeed.tech/tags/results.md>), [social](<https://devfeed.tech/tags/social.md>), [workshop](<https://devfeed.tech/tags/workshop.md>)

### AI overview

At Lumen Technologies' Innovation Summit 2026, 31 network engineers ran 40 distinct Itential FlowAgents during the workshop's busiest 45 minutes. They explored device health, inventory discovery, configuration changes, and drift audits, progressing from individual agents to spec-driven automation and coordinated agent teams. The post describes a governed execution model: AI agents interpret intent and determine next steps, while deterministic workflows execute actions through approvals, repeatable steps, and an audit trail. It reports workshop activity and the approach explored, not measured operational outcomes.

### Source excerpt

Itential - LinkedIn - Itential shares the workshop results and recognizes the Lumen engineers and partners who explored agents, specifications, and governed execution together.

## Marc Montez: AI decides, workflows execute

DevFeed: [Marc Montez: AI decides, workflows execute](<https://devfeed.tech/articles/marc-montez-ai-decides-workflows-execute-110107.md>)

Original publisher: [Read original article](<https://www.linkedin.com/posts/marc-montez-227b138_networkautomation-aiagents-itential-activity-7519521159781212161-yxcK>)

Author: Marc Montez

Published: 2026-10-23T12:00:00Z

Content type: opinion

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>)

Tags: [a2a](<https://devfeed.tech/tags/a2a.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [audit](<https://devfeed.tech/tags/audit.md>), [audit-trail](<https://devfeed.tech/tags/audit-trail.md>), [automation](<https://devfeed.tech/tags/automation.md>), [control](<https://devfeed.tech/tags/control.md>), [deterministic-workflows](<https://devfeed.tech/tags/deterministic-workflows.md>), [execution](<https://devfeed.tech/tags/execution.md>), [guardrails](<https://devfeed.tech/tags/guardrails.md>), [itential](<https://devfeed.tech/tags/itential.md>), [john-capobianco](<https://devfeed.tech/tags/john-capobianco.md>), [lumen](<https://devfeed.tech/tags/lumen.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [network](<https://devfeed.tech/tags/network.md>), [partnership](<https://devfeed.tech/tags/partnership.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [social](<https://devfeed.tech/tags/social.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

Marc Montez argues that production network automation should pair AI agent reasoning--interpreting intent, correlating signals, and deciding next steps--with deterministic workflows that execute changes predictably and repeatably. He says those workflows provide guardrails, approvals, observability, and an audit trail, aiming for speed while retaining control and trust. He cites work with Itential and says they are exploring MCP and agent-to-agent patterns for production-grade orchestration.

### Source excerpt

Marc Montez - LinkedIn - Marc describes the partnership between agent reasoning and deterministic workflows, with approvals, repeatable execution, and visible evidence.

## Agentic ServiceNow Remediation: From Alert to Closed Ticket with FlowAgents

DevFeed: [Agentic ServiceNow Remediation: From Alert to Closed Ticket with FlowAgents](<https://devfeed.tech/articles/agentic-servicenow-remediation-from-alert-to-closed-ticket-with-flowagents-110110.md>)

Original publisher: [Read original article](<https://www.youtube.com/watch?v=Dq1YhZ602kA>)

Published: 2026-10-22T12:00:00Z

Content type: tutorial

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai-agents-for-network-operations](<https://devfeed.tech/tags/ai-agents-for-network-operations.md>), [alert](<https://devfeed.tech/tags/alert.md>), [appearance](<https://devfeed.tech/tags/appearance.md>), [documentation](<https://devfeed.tech/tags/documentation.md>), [flowagents](<https://devfeed.tech/tags/flowagents.md>), [human-in-the-loop](<https://devfeed.tech/tags/human-in-the-loop.md>), [incident](<https://devfeed.tech/tags/incident.md>), [incident-remediation](<https://devfeed.tech/tags/incident-remediation.md>), [itential](<https://devfeed.tech/tags/itential.md>), [itential-flowai](<https://devfeed.tech/tags/itential-flowai.md>), [john-capobianco](<https://devfeed.tech/tags/john-capobianco.md>), [joksan-flores](<https://devfeed.tech/tags/joksan-flores.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [network-troubleshooting-automation](<https://devfeed.tech/tags/network-troubleshooting-automation.md>), [remediation](<https://devfeed.tech/tags/remediation.md>), [servicenow](<https://devfeed.tech/tags/servicenow.md>), [servicenow-ai-agents](<https://devfeed.tech/tags/servicenow-ai-agents.md>), [servicenow-automation](<https://devfeed.tech/tags/servicenow-automation.md>), [ticket](<https://devfeed.tech/tags/ticket.md>), [triage](<https://devfeed.tech/tags/triage.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

John Capobianco and Joksan Flores demonstrate a ServiceNow incident workflow using FlowAgents: coordinate the incident, perform read-only triage, obtain human approval before remediation, and document the result back in ServiceNow.

### Source excerpt

Itential - Guest appearance - John Capobianco and Joksan Flores demonstrate incident coordination, read-only triage, human-approved remediation, and documentation back in ServiceNow.

## Greg Freeman: a summit built around the engineers

DevFeed: [Greg Freeman: a summit built around the engineers](<https://devfeed.tech/articles/greg-freeman-a-summit-built-around-the-engineers-110103.md>)

Original publisher: [Read original article](<https://www.linkedin.com/posts/greg-freeman-9311521_ai-networkautomation-lumen-activity-7519125720619536385-uKHb>)

Author: Greg Freeman

Published: 2026-10-22T12:00:00Z

Content type: opinion

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Developer Events](<https://devfeed.tech/topics/developer-events.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [collaboration](<https://devfeed.tech/tags/collaboration.md>), [event](<https://devfeed.tech/tags/event.md>), [innovation](<https://devfeed.tech/tags/innovation.md>), [itential](<https://devfeed.tech/tags/itential.md>), [lumen](<https://devfeed.tech/tags/lumen.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [partners](<https://devfeed.tech/tags/partners.md>), [phoenix](<https://devfeed.tech/tags/phoenix.md>), [presentations](<https://devfeed.tech/tags/presentations.md>), [science](<https://devfeed.tech/tags/science.md>), [sessions](<https://devfeed.tech/tags/sessions.md>), [social](<https://devfeed.tech/tags/social.md>), [summit](<https://devfeed.tech/tags/summit.md>), [tools](<https://devfeed.tech/tags/tools.md>), [worlds](<https://devfeed.tech/tags/worlds.md>)

### AI overview

Lumen's Phoenix AI Innovation Summit brought together more than 200 engineers for three days, with main-stage presentations, 45 breakout sessions, and hands-on "Science Fair" kiosks featuring custom builds with partner tools including Anthropic and Itential. Greg Freeman's recap emphasizes building and collaboration alongside discussion of network automation and AI. The excerpt does not describe the demonstrations' technical details.

### Source excerpt

Greg Freeman - LinkedIn - Greg's recap of Lumen's three-day summit: 200+ engineers, 45 breakout sessions, and hands-on demonstrations supported by partners including Itential.

## Jev Told My FlowAgents to Roll Back. I Kept That in the Demo.

DevFeed: [Jev Told My FlowAgents to Roll Back. I Kept That in the Demo.](<https://devfeed.tech/articles/jev-told-my-flowagents-to-roll-back-i-kept-that-in-the-demo-110100.md>)

Original publisher: [Read original article](<https://www.itential.com/resource/blog/jev-decision-models-flowagents-network-changes/>)

Author: John Capobianco

Published: 2026-10-20T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [decision-making](<https://devfeed.tech/topics/decision-making.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [function calling](<https://devfeed.tech/topics/function-calling.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [about](<https://devfeed.tech/tags/about.md>), [act](<https://devfeed.tech/tags/act.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-harness](<https://devfeed.tech/tags/agent-harness.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-reasoning](<https://devfeed.tech/tags/ai-reasoning.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [api](<https://devfeed.tech/tags/api.md>), [article](<https://devfeed.tech/tags/article.md>), [claude](<https://devfeed.tech/tags/claude.md>), [gemma](<https://devfeed.tech/tags/gemma.md>), [i](<https://devfeed.tech/tags/i.md>), [itential](<https://devfeed.tech/tags/itential.md>), [jev](<https://devfeed.tech/tags/jev.md>), [my](<https://devfeed.tech/tags/my.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [tools](<https://devfeed.tech/tags/tools.md>)

### AI overview

A network change workflow used Claude Opus to coordinate, Sonnet agents to investigate, plan, execute and validate, Gemma to report, and Jev to assess evidence through an MCP wrapper connected via Itential Gateway. Scoped tools limited investigators to show commands and gave executors configuration and testing capabilities. In a human approval mode, Work Center presented proposed commands, checks, rollback plan, and Jev's assessments for review. In an earlier run, three of 14 checks failed; Jev rejected the result and requested rollback, but recovery was incomplete and needed operator attention. Later runs passed 30 plan checks and, respectively, 15 result checks or the named checks described for the approved run; Jev accepted the outcomes. One run completed without a human approval task or rollback. The author stresses that confidence scores do not replace inspection of commands and test evidence, and that rollback requests and successful recovery are distinct outcomes.

### Source excerpt

Itential - Article - A demonstration of decision gates, scoped tools, human approvals, and recovery: Claude coordinates, Jev evaluates evidence, and Gemma handles reporting.

## The Network Engineer Who Hates Programming Just Built an App

DevFeed: [The Network Engineer Who Hates Programming Just Built an App](<https://devfeed.tech/articles/the-network-engineer-who-hates-programming-just-built-an-app-110108.md>)

Original publisher: [Read original article](<https://www.linkedin.com/pulse/network-engineer-who-hates-programming-just-built-app-john-capobianco-2a9hc>)

Author: John Capobianco

Published: 2026-10-18T12:00:00Z

Content type: opinion

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [Programming](<https://devfeed.tech/topics/programming.md>), [Test-driven development](<https://devfeed.tech/topics/tdd.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Learning](<https://devfeed.tech/topics/learning.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [acceptance](<https://devfeed.tech/tags/acceptance.md>), [accessible](<https://devfeed.tech/tags/accessible.md>), [advice](<https://devfeed.tech/tags/advice.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [all](<https://devfeed.tech/tags/all.md>), [app](<https://devfeed.tech/tags/app.md>), [app-store](<https://devfeed.tech/tags/app-store.md>), [article](<https://devfeed.tech/tags/article.md>), [building](<https://devfeed.tech/tags/building.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [just](<https://devfeed.tech/tags/just.md>), [learning](<https://devfeed.tech/tags/learning.md>), [links](<https://devfeed.tech/tags/links.md>), [network](<https://devfeed.tech/tags/network.md>), [network-engineer](<https://devfeed.tech/tags/network-engineer.md>), [programming](<https://devfeed.tech/tags/programming.md>), [public](<https://devfeed.tech/tags/public.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

A network engineer who disliked programming built the AONE Mobile app after learning to guide AI through a structured workflow. The article traces his progress from a hands-on Hello World using NetBox's MCP server--where setup errors and an agent's mistaken approach required troubleshooting and clearer prompts--to Spec-Driven Development with GitHub Spec Kit and Claude Code. Its six steps are Constitution, Specify, Clarify, Plan, Tasks and Implement; planning artifacts stay in Git, and the AI asks questions before coding. The article reports 77 passing tests and full subnetting-engine coverage in the demonstration. The engineer later applied the method to the app, writing its specs himself; its constitution requires verifying features on a real device. The article recommends starting with a bounded side project, doing the setup yourself, and checking the AI's output. AONE Mobile is described as an iPhone and Apple Watch companion whose content largely reads from the show's public GitHub repository; it was described as coming soon to the App Store.

### Source excerpt

LinkedIn - Article - John follows Andy Lapteff's public learning journey from initial AI experiments to building an app, with links to their conversations and demonstrations.

## Two Network Geeks, One Table: Eva Santos in the VibeOps Lounge

DevFeed: [Two Network Geeks, One Table: Eva Santos in the VibeOps Lounge](<https://devfeed.tech/articles/two-network-geeks-one-table-eva-santos-in-the-vibeops-lounge-88894.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/ai/vibeops-lounge-eva-santos-chromeos-wifi/>)

Author: John Capobianco

Published: 2026-10-16T12:00:00Z

Content type: article

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [AI in frontend development](<https://devfeed.tech/topics/ai-in-frontend-development.md>), [Cloud Networking](<https://devfeed.tech/topics/cloud-networking.md>), [Endpoint security](<https://devfeed.tech/topics/endpoint-security.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [article](<https://devfeed.tech/tags/article.md>), [chromeos](<https://devfeed.tech/tags/chromeos.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [community](<https://devfeed.tech/tags/community.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [network-automation](<https://devfeed.tech/tags/network-automation.md>), [vibeops](<https://devfeed.tech/tags/vibeops.md>), [wi-fi](<https://devfeed.tech/tags/wi-fi.md>), [wireshark](<https://devfeed.tech/tags/wireshark.md>)

### AI overview

An episode of VibeOps Lounge features wireless engineer Eva Santos discussing Wi-Fi troubleshooting, including a deauthentication issue diagnosed with a packet capture and rogue containment that disrupted a city bus Wi-Fi network. Santos demonstrates ChromeOS Wi-Fi Fixer, which turns raw diagnostics into readable connection timelines and signal data, and describes building it with Claude Code. The conversation also covers finding recurring engineering problems that can be turned into reusable software tools.

### Source excerpt

From Walmart freight to CWNE #521: Eva Santos joins me to talk Wireshark, ChromeOS Wi-Fi Fixer, Claude Code, and building useful tools before the token buffet closes.

[Next page](<https://devfeed.tech/sources/automate-your-network.md?cursor=WyIyMDI2LTEwLTE2VDEyOjAwOjAwKzAwOjAwIiwgIjYxNWM1NWE0LWM3NGMtNDFmOC04Nzg3LWUwNTI4OWNkZGIyMyJd>)