# Why the best agentic security removes engineering friction | LinearB Blog

> 1Password CTO Nancy Wang on why the best agentic security removes engineering friction. Discover how no long-lived secrets keep agents and engineers safe.

_This is a markdown rendering of a live HTML page on linearb.io, generated for AI/LLM consumption — it is not a markdown-only site. To get the full HTML page instead, request this URL with an explicit `Accept: text/html` header (no wildcard, no markdown preference)._


```json
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    {
      "@type": "ListItem",
      "position": 1,
      "name": "Home",
      "item": "https://linearb.io/"
    },
    {
      "@type": "ListItem",
      "position": 2,
      "name": "Blog",
      "item": "https://linearb.io/blog"
    },
    {
      "@type": "ListItem",
      "position": 3,
      "name": "Why the best agentic security removes engineering friction",
      "item": "https://linearb.io/blog/1password-nancy-wang-agentic-security-no-long-lived-secrets"
    }
  ]
}
```

[Home](https://linearb.io/)

/

[Blog](https://linearb.io/blog)

/

Why the best agentic security removes engineering friction

# Why the best agentic security removes engineering friction

![Photo of Andrew Zigler](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Headshot3_d7231cbda7?_a=BAVMn6ID0)

By [Andrew Zigler](https://linearb.io/blog/1password-nancy-wang-agentic-security-no-long-lived-secrets#andrew-zigler)

|

August 28, 2026

![Blog_Whythebestagenticsecurityremoves_2400x1256_e31ce7ae58](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Blog_Whythebestagenticsecurityremoves_2400x1256_e31ce7ae58?_a=BAVMn6ID0)

The first documented end-to-end autonomous ransomware attack did not begin with a rogue model. It began the way breaches almost always begin, with a human and a stolen credential. That detail sits at the center of how [Nancy Wang, CTO at 1Password](https://linearb.io/dev-interrupted/podcast/1password-nancy-wang-agentic-security-secrets), thinks about the security models now colliding with agentic development. Before 1Password, she built infrastructure at Google and AWS, and that systems-first instinct runs through everything her team ships, from background agents in the back end to a credential broker built for a world where machines may soon outnumber people at the vault.

Under her leadership, 1Password has rolled agentic workflows into some of the toughest parts of its own stack while holding a zero knowledge guarantee it refuses to bend. The result is a working thesis for keeping both humans and agents secure without slowing engineers down. It is a thesis grounded less in novelty than in fundamentals, the kind that were true before agents and are only more urgent now.

## AI agent security demands rethinking trust after human error

JadePuffer, the AI-driven ransomware attack disclosed by researchers at Sysdig in July 2026, is the cautionary tale that reframes the whole conversation. The attack ran end to end autonomously once it got inside, yet as Wang points out, it still started with a human using stolen credentials. An agent got in because a person's credentials were sitting where an attacker could take them, and then it ran the ransomware playbook. The autonomy was new, but the entry point was as old as security itself.

That distinction matters because so much agentic risk discourse fixates on the model going rogue while ignoring the delegated identity acting inside systems built for humans. To understand how these tools behave, 1Password dogfoods aggressively across Claude Code, Codex, Copilot, Cursor, and more, because a company promising security for coding agents has to know how those agents reach across systems, make tool calls, and touch production. Security models designed for human operators now have to account for agents acting as stand-ins for those operators, which changes what trust has to mean at runtime.

## No long-lived secrets eliminate the agent's attack surface

The countermeasure is a discipline security has preached forever and rarely enforced cleanly, which is no long-lived secrets. The recent OpenAI Codex beta puts the idea into practice. A binary ships with the local 1Password desktop app, and whenever Codex wants to connect to a system or make a tool call, the raw secret never lands in a prompt window, a file, a terminal, or the model context.

The agent still orchestrates, reasoning and staying creative while the app executes. What it never receives is the secret. Wang calls this "access without custody." The credential is used at the point of execution and never handed over, so there is nothing to steal from a prompt, a file, or a terminal buffer.

This builds on the zero knowledge encryption model underneath the entire product. Credentials sit behind asymmetric key encryption, and even 1Password cannot read them. Only the customer holds the private key, and anyone who did get inside a vault would find encrypted blobs. If agents cannot see your secrets, it is because the company itself cannot. Standing secrets are exactly the gap JadePuffer exploited, and removing them removes the surface area an attacker was counting on.

## Secrets management for coding agents keeps credentials out of reach

Eliminating standing secrets raises the obvious question of how anything gets access at all. The answer is a credential broker that retires the old static service account. Once an identity is proven, a coding agent can declare a goal, receive exactly what the task needs, and lose access the moment the task ends. Oracle Red Bull shows the pattern in production. The team uses 1Password as a centralized secrets store, then uses the Kubernetes operator and connect server to inject secrets at runtime. The rule, in Wang's words, is "in running processes, never on disk." Nothing sits in a stray config file waiting to be discovered, and the credential exists only inside the live process that requires it.

A wind tunnel recovery workflow that took roughly 60 minutes dropped to under five with an Ansible playbook, without loosening a single control. The Codex integration runs on the same principle, injecting values only into an authorized runtime process and handing them to nothing else. Secrets [management for coding agents](https://linearb.io/resources/software-factory) works when the secret is out of reach, sitting in a cookie jar on a shelf the agent cannot climb to.

## Security as the paved path makes engineering faster

None of this works if it slows engineers down, and done right it does the opposite. The old ritual sent a PR out and waited for security to run its check. The better model folds that check into CI or a postscript step, as with the Cursor Hooks launch, so the secure path stops being a gate and starts being the default. "Let the paved path be the easiest and the fastest path," she says. Baking controls in from the start makes engineering faster.

The best security tooling removes friction. It kills the weird workarounds engineers invent to route around a checkpoint, which are usually where risk creeps back in. [That philosophy now lives inside code review itself](http://linearb.io/blog/best-ai-code-review-tool-benchmark-linearb). An upcoming report details an internal security harness that distills more than 300 rules 1Password has used since its earliest days into an AI-assisted review program. Engineers no longer wait on tickets from the AppSec team. The harness runs in the review process, it is self-learning, and it cuts time to value for everyone touching the code.

The same discipline shapes how she judges the acceleration. Code is cheap now, and what has become expensive is review, security review, and deployment, so PR counts moving from 2.5 a week to 3.5 are a weak signal. [A board conversation turns on features delivered](http://linearb.io/blog/cto-cfo-ai-core-business-metric), and the anecdote she hears from CTO peers is a team clearing two or three items off a backlog that was never staffed for the quarter. Features built almost entirely with agents still answer to the same four-nines uptime bar, the same full test suite, and the same manual pass.

## Agent identity and non-human identities redefine access control

All of this rests on a single reframe of what an identity is. 1Password designs for a world where humans, machines, and agents are equally first-class at the vault, and where non-human identities may soon outnumber the human ones.

That shift turns identity into a runtime problem. Caching a secret and handing it over is no longer enough. Access control has to know, in real time, who the agent is, which human delegated to it, what goal it declared, which tools are in scope, how long the token should live, and what policy governs the action at that exact moment. The Apono acquisition brings just-in-time, zero-standing-privilege access governance for human, machine, and agent identities directly into that runtime picture, because deciding whether to grant an agent access, for that goal, under that policy, in that instant, is the new shape of the problem.

The most sensitive decisions never fully automate. [Agents are best at deterministic toil](http://linearb.io/blog/ai-coding-agents-code-refactoring), where success is easy to test and rollbacks are easy to run. Ambiguous judgment, the policy engine itself, anything touching production, anything crossing a customer trust boundary, and anything approaching a one-way door all keep a human in the loop. That is what keeps non-human identities accountable.

## Agents made the old security textbooks mandatory

The through line is that agents did not make the old textbooks obsolete. They made them required reading. Wang is blunt about the timing. "Now is exactly the right time to do zero trust. Now is exactly the right time to do identity verification." Scoped access, human sign-off on one-way doors, and golden copies of sensitive data stored somewhere an attacker cannot reach are the fundamentals that let a security company [ship features almost entirely with agents](http://linearb.io/blog/sdlc-to-adlc-agentic-software-delivery-transformation) without ever compromising a zero knowledge guarantee.

The paved path and the fastest path can be the same path, but only for teams disciplined enough to keep the secret out of the model's hands and the human in the loop where it counts. The friction was never the security. It was the workaround.

To hear more of Nancy Wang's insights on no long-lived secrets, zero trust for autonomous agents, and agent identity, listen to the full episode on the Dev Interrupted podcast.

## Improve developer productivity with LinearB

Find us on

[](https://www.linkedin.com/company/linearb)
[](https://devinterrupted.substack.com/)

![Headshot3_d7231cbda7](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Headshot3_d7231cbda7?_a=BAVMn6ID0)

## Andrew Zigler

Andrew Zigler is a GTM Engineer at LinearB and the host of Dev Interrupted, a twice-weekly podcast and newsletter where 40k+ builders decode the transition to AI-native development and agentic orchestration. A classicist by training with a degree from The University of Texas at Austin, Andrew spent his early career teaching in Japan before channeling his interdisciplinary instincts into the tech world. His polymath background informs everything he builds, from automated workflows to the stories he tells about the seismic shifts reshaping software creation.

### Connect with

[](https://www.linkedin.com/in/andrewzigler)
[](https://substack.com/@zigler)
[](https://x.com/andrewzigler)

## Your next read

[![Cover image for How to keep seeing when your software factory runs in the dark](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Blog_Post_Name_2400x1256_5_c9af1d9dc4?_a=BAVMn6ID0)](https://linearb.io/blog/software-factory-night-vision)

AI

[How to keep seeing when your software factory runs in the dark](https://linearb.io/blog/software-factory-night-vision)

How to keep visibility as your SDLC runs without humans: LinearB, HumanLayer and Warp on tracking autonomy and measuring quality past the merge. 

[![Cover image for Simplicity in model-driven agents and MCP drives real engineering gains](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Blog_Simplicity_in_model_driven_agents_2400x1256_e67f24d00b?_a=BAVMn6ID0)](https://linearb.io/blog/aws-clare-liguori-model-driven-agents-mcp-simplicity)

AI

[Simplicity in model-driven agents and MCP drives real engineering gains](https://linearb.io/blog/aws-clare-liguori-model-driven-agents-mcp-simplicity)

AWS's Clare Liguori on why the model, not the scaffolding, should carry the load in agent development. Discover how simpler agents and MCP drive real gains.

[![Cover image for Elite AI-driven engineering teams double PR output while others stall](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Blog_Elite_AI_driven_Teams_2400x1256_fd33821a9d?_a=BAVMn6ID0)](https://linearb.io/blog/elite-engineering-teams-double-pr-output-ai-benchmarks)

AI

[Elite AI-driven engineering teams double PR output while others stall](https://linearb.io/blog/elite-engineering-teams-double-pr-output-ai-benchmarks)

LinearB CTO Yishai Beeri breaks down benchmark data revealing how elite AI-driven engineering teams doubled their merged PR output while others stalled. Learn...

## Structured data

_Machine-readable metadata (JSON-LD) embedded in the page for search/AI context — not content rendered on the page itself._

```json
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "LinearB",
  "url": "https://linearb.io/",
  "logo": "https://assets.linearb.io/image/upload/v1715628027/logo-mark-lg.svg",
  "description": "LinearB is the engineering productivity platform that helps engineering leaders prove AI is improving throughput without sacrificing delivery confidence, flow efficiency, or developer experience.",
  "sameAs": [
    "https://www.linkedin.com/company/linearb"
  ],
  "award": [
    {
      "@type": "Award",
      "name": "LinearB is a Leader in the 2026 Gartner® Magic Quadrant™ for Developer Productivity Insight Platforms",
      "dateAwarded": "2026",
      "awardedBy": {
        "@type": "Organization",
        "name": "Gartner®"
      }
    },
    {
      "@type": "Award",
      "name": "Great Place to Work Certification",
      "dateAwarded": "2025-2027",
      "awardedBy": {
        "@type": "Organization",
        "name": "Great Place to Work"
      }
    },
    {
      "@type": "Award",
      "name": "America's Best Startup Employers 2025",
      "dateAwarded": "2025",
      "awardedBy": {
        "@type": "Organization",
        "name": "Forbes Magazine"
      }
    }
  ],
  "hasCertification": [
    {
      "@type": "Certification",
      "name": "SOC 1 Type 2"
    },
    {
      "@type": "Certification",
      "name": "SOC 2 Type 2"
    },
    {
      "@type": "Certification",
      "name": "GDPR Compliance certification"
    },
    {
      "@type": "Certification",
      "name": "ISO 27001"
    }
  ]
}
```

```json
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Why the best agentic security removes engineering friction",
  "url": "https://linearb.io/blog/1password-nancy-wang-agentic-security-no-long-lived-secrets",
  "author": {
    "@type": "Person",
    "name": "Andrew Zigler"
  },
  "datePublished": "2026-08-28T07:00:00.000Z",
  "dateModified": "2026-08-28T07:00:00.000Z",
  "image": "https://assets.linearb.io/image/upload/v1788897930/Blog_Whythebestagenticsecurityremoves_2400x1256_e31ce7ae58.png",
  "publisher": {
    "@type": "Organization",
    "name": "LinearB",
    "logo": "https://assets.linearb.io/image/upload/v1777485755/linearb-logo-2026.png"
  },
  "description": "1Password CTO Nancy Wang on why the best agentic security removes engineering friction. Discover how no long-lived secrets keep agents and engineers safe."
}
```

## More on linearb.io

### Top navigation

- [Book a Demo](https://linearb.io/book-a-demo)
- [AI Code Reviews — Catch security risks, bugs, and spec mismatches](https://linearb.io/platform/ai-code-reviews)
- [AI & Productivity Insights — See how AI tools affect cycle time and delivery speed](https://linearb.io/platform/ai-developer-productivity-insights)
- [Measure AI Impact — Track AI adoption and tie it to delivery outcomes](https://linearb.io/use-case/measure-ai-impact)
- [MCP Server — Chat with your data to spot patterns and boost output](https://linearb.io/platform/mcp-server)
- [Resource Allocation — Cost initiatives and shape your investment strategy](https://linearb.io/platform/resource-allocation)
- [Cost Capitalization — Capitalize engineering costs with audit-ready reports](https://linearb.io/platform/cost-capitalization)
- [Dev Team Management — Set targets and tie throughput to business outcomes](https://linearb.io/platform/goals-and-reporting)
- [DevOps Workflow Automation — Policy-based PR routing, approvals, and tests](https://linearb.io/platform/ai-workflow-governance)
- [AI Powered Support — Unify AI and human code delivery in one clear view](https://linearb.io/use-case/ai-powered-support)
- [Optimization — Surface friction with feedback and MCP insights](https://linearb.io/platform/developer-experience)
- [Reporting — Spot what's working and what needs attention](https://linearb.io/use-case/measuring-developer-experience)
- [Surveys — Turn developer feedback into actionable signals](https://linearb.io/platform/developer-surveys)
- [Platform overview](https://linearb.io/platform/overview)
- [Watch now](https://linearb.io/resources/the-great-software-factory-debate)
- [Customers](https://linearb.io/customers)
- [Pricing](https://linearb.io/pricing)
- [Why choose LinearB — Explore your data. Measure performance. Act to improve it.](https://linearb.io/why-linearb)
- [APEX framework — The operating model for AI-era engineering teams](https://linearb.io/resources/apex-framework)
- [Anti-FAQ — The questions other vendors won't answer](https://linearb.io/why-linearb/anti-faq)
- [Security — Enterprise-grade compliance and zero code access](https://linearb.io/security)
- [Build vs. buy — The hidden cost of building it yourself](https://linearb.io/resources/build-vs-buy)
- [Dev Interrupted Podcast — Conversations with engineering leaders](https://linearb.io/dev-interrupted/podcasts)
- [Reports & Guides — Deep dives on productivity and delivery](https://linearb.io/resources)
- [Webinars — Expert sessions on productivity and AI](https://linearb.io/resources?category=workshops)
- [Metrics Benchmarks — See how your engineering org stacks up](https://linearb.io/resources/software-engineering-benchmarks-report)
- [Blog — Product updates and practical insights](https://linearb.io/blog)
- [Help Center — Documentation, setup, and support](https://linearb.helpdocs.io)
- [API Docs](https://docs.linearb.io/api-overview)
- [Status](https://www.linearbstatus.com/)
- [Integrations](https://linearb.io/integrations)
- [LinearB Library](https://linearb.io/library)
- [Engineering metrics](https://linearb.io/library/engineering-metrics)
- [Platform engineering](https://linearb.io/library/platform-engineering)
- [Engineering glossary](https://linearb.io/library/engineering-glossary)
- [Developer productivity](https://linearb.io/library/developer-productivity)
- [AI in software development](https://linearb.io/library/ai-in-software-development)
- [Engineering management](https://linearb.io/library/engineering-management)
- [Developer experience](https://linearb.io/library/developer-experience)
- [DevOps](https://linearb.io/library/devops)
- [Engineering operations and the context layer](https://linearb.io/library/engineering-operations)
- [Engineering efficiency](https://linearb.io/library/engineering-efficiency)
- [Software delivery](https://linearb.io/library/software-delivery)
- [Research and data](https://linearb.io/library/engineering-benchmarks-and-research)
- [LinearB is a Leader in the 2026 Gartner® Magic Quadrant™ for Developer Productivity Insight Platforms](https://linearb.io/resources/gartner-magic-quadrant-dpi-platforms-2026)
- [Sign in](https://app.linearb.io/login)
- [Enterprise](https://linearb.io/solutions/enterprise)
- [Contact](https://linearb.io/contact-us)
- [About us](https://linearb.io/about-us)
- [Careers](https://linearb.io/careers)
- [Service agreement](https://linearb.io/services-agreement)
- [Privacy policy](https://linearb.io/privacy-policy)
- [DPA](https://linearb.io/data-processing-agreement)
- [Security FAQ](https://linearb.io/security-faq)
- [Substack](https://devinterrupted.substack.com/)

### Footer

_Additional links from the site footer, not repeated from the top navigation above._

- [GitHub](https://github.com/linear-b)
- [LinkedIn](https://www.linkedin.com/company/linearb)
- [Twitter](https://twitter.com/LinearB_Inc)