--- title: Anthropic built an inspection layer that lets... description: Anthropic launched inference hooks for Claude Enterprise, a security feature that routes every prompt and tool call through a customer's own DLP (Data Loss... canonical: https://daily.dev/posts/anthropic-built-an-inspection-layer-that-lets-enterprises-block-sensitive-data-before-it-reaches-cla-oy4teifgp twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Anthropic built an inspection layer that lets enterprises block sensitive data before it reaches Claude | daily.dev og:description: Anthropic launched inference hooks for Claude Enterprise, a security feature that routes every prompt and tool call through a customer's own DLP (Data Loss... og:url: https://daily.dev/posts/anthropic-built-an-inspection-layer-that-lets-enterprises-block-sensitive-data-before-it-reaches-cla-oy4teifgp og:image: https://api.daily.dev/og/posts/OY4tEiFGP.png og:image:alt: Anthropic built an inspection layer that lets enterprises block sensitive data before it reaches Claude og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Anthropic built an inspection layer that lets enterprises block sensitive data before it reaches Claude **[The Next Web](https://daily.dev/sources/tnw)** · 2 min read · 1 upvotes · 0 comments ## Summary Anthropic launched inference hooks for Claude Enterprise, a security feature that routes every prompt and tool call through a customer's own DLP (Data Loss Prevention) server before Claude processes it. The server returns an allow or deny verdict, and Claude only proceeds if allowed. The same inspection applies to tool call responses via MCP, skills, and plugins. One configuration covers all Claude surfaces — chat, Claude Code, Cowork, and connected tools. It integrates with existing DLP infrastructure from Netskope, Palo Alto Networks, Proofpoint, and Zscaler, or custom servers. Features include shadow mode for monitoring before enforcement, role-based exclusions, and percentage-based rollouts for gradual deployment. Available in beta for Claude Enterprise customers. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is Anthropic's inference hooks feature for Claude Enterprise? Inference hooks is a beta feature that routes every prompt and tool call through a customer's own data loss prevention server before Claude processes it, with the DLP server returning an allow or deny verdict. A single organization-level configuration covers chat, Claude Code, Cowork, MCP connectors, and plugins, replacing per-product integrations previously required. _Enterprises evaluating AI data governance can follow Claude platform changes like this on daily.dev._ ### Which DLP vendors work with Anthropic's inference hooks? Inference hooks uses a webhook-based protocol with a published schema designed to integrate with Netskope, Palo Alto Networks, Proofpoint, Zscaler, or custom-built security servers. Organizations can also run shadow mode, which always allows requests while logging them, or phase in enforcement using role-based exclusions and percentage-based rollouts. _Teams choosing a DLP integration path for Claude can track vendor compatibility updates on daily.dev._ ### How did Claude Code hooks limit enterprise security before inference hooks launched? Before inference hooks, inline enforcement of data controls was limited to Claude Code's client-side hooks, leaving chat, Cowork, and connected tool surfaces without a unified inspection point. Inference hooks extends enforcement to every Claude Enterprise surface with a single organization-level configuration instead of separate integrations per product. _Security teams standardizing controls across AI surfaces can follow this shift on daily.dev._ ## Similar posts on daily.dev - [Announcing Claude Compliance API support with Cloudflare CASB](https://daily.dev/posts/announcing-claude-compliance-api-support-with-cloudflare-casb-ggphkdqdk) · Cloudflare · 0 upvotes · 0 comments - [Claude's collaboration tools allowed remote code execution](https://daily.dev/posts/claude-s-collaboration-tools-allowed-remote-code-execution-jt8sizuec) · The Register · 0 upvotes · 0 comments - [Claude Code leak puts enterprise trust at risk as security, governance concerns mount](https://daily.dev/posts/claude-code-leak-puts-enterprise-trust-at-risk-as-security-governance-concerns-mount-nf4klga2o) · InfoWorld · 1 upvotes · 0 comments --- Tags: [#security](https://daily.dev/tags/security), [#claude](https://daily.dev/tags/claude), [#anthropic](https://daily.dev/tags/anthropic) [View this post on daily.dev](https://daily.dev/posts/anthropic-built-an-inspection-layer-that-lets-enterprises-block-sensitive-data-before-it-reaches-cla-oy4teifgp) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Anthropic built an inspection layer that lets enterprises block sensitive data before it reaches Claude","url":"https://daily.dev/posts/anthropic-built-an-inspection-layer-that-lets-enterprises-block-sensitive-data-before-it-reaches-cla-oy4teifgp","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/anthropic-built-an-inspection-layer-that-lets-enterprises-block-sensitive-data-before-it-reaches-cla-oy4teifgp"},"datePublished":"2026-08-05T18:50:59.581Z","dateModified":"2026-09-14T06:54:21.688Z","description":"Anthropic launched inference hooks for Claude Enterprise, a security feature that routes every prompt and tool call through a customer's own DLP (Data Loss...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b6ffa4fe175041b5a5918a3aeceeec80?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b6ffa4fe175041b5a5918a3aeceeec80?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The Next Web","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"The Next Web","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/tnw","url":"https://daily.dev/sources/tnw"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/anthropic-built-an-inspection-layer-that-lets-enterprises-block-sensitive-data-before-it-reaches-cla-oy4teifgp","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"security,claude,anthropic","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The Next Web","item":"https://daily.dev/sources/tnw"},{"@type":"ListItem","position":3,"name":"Anthropic built an inspection layer that lets enterprises block sensitive data before it reaches Claude"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/anthropic-built-an-inspection-layer-that-lets-enterprises-block-sensitive-data-before-it-reaches-cla-oy4teifgp#faq","mainEntity":[{"@type":"Question","name":"What is Anthropic's inference hooks feature for Claude Enterprise?","acceptedAnswer":{"@type":"Answer","text":"Inference hooks is a beta feature that routes every prompt and tool call through a customer's own data loss prevention server before Claude processes it, with the DLP server returning an allow or deny verdict. A single organization-level configuration covers chat, Claude Code, Cowork, MCP connectors, and plugins, replacing per-product integrations previously required. Enterprises evaluating AI data governance can follow Claude platform changes like this on daily.dev."}},{"@type":"Question","name":"Which DLP vendors work with Anthropic's inference hooks?","acceptedAnswer":{"@type":"Answer","text":"Inference hooks uses a webhook-based protocol with a published schema designed to integrate with Netskope, Palo Alto Networks, Proofpoint, Zscaler, or custom-built security servers. Organizations can also run shadow mode, which always allows requests while logging them, or phase in enforcement using role-based exclusions and percentage-based rollouts. Teams choosing a DLP integration path for Claude can track vendor compatibility updates on daily.dev."}},{"@type":"Question","name":"How did Claude Code hooks limit enterprise security before inference hooks launched?","acceptedAnswer":{"@type":"Answer","text":"Before inference hooks, inline enforcement of data controls was limited to Claude Code's client-side hooks, leaving chat, Cowork, and connected tool surfaces without a unified inspection point. Inference hooks extends enforcement to every Claude Enterprise surface with a single organization-level configuration instead of separate integrations per product. Security teams standardizing controls across AI surfaces can follow this shift on daily.dev."}}]} ```