Inside Claude Managed Agents: Reverse-Engineering the Security Boundaries of Anthropic's Hosted Agent Runtime

This Pluto Security blog post reverse-engineers Anthropic's Claude Managed Agents (a hosted runtime where Claude runs autonomously in cloud containers with bash, file I/O, web access, and MCP tools). Key findings include: the sandbox uses gVisor with a three-layer egress control system (the same isolation engine as Claude Cowork); all outbound traffic routes through a JWT-authenticated egress proxy with TLS inspection; the JWT is readable by any process in the sandbox and reveals organization metadata, session ID, and allowed hosts; even in "limited" networking mode, six additional Anthropic infrastructure hosts (including sentry.io and a staging endpoint) are silently injected into the egress JWT beyond user configuration. Three independent layers prevent proxy bypass (no DNS, network firewall, JWT validation). The vault credential proxy is identified as the platform's strongest security property — vault secrets never enter the sandbox, structurally preventing credential theft via prompt injection. However, the default configuration is maximally permissive (all eight tools enabled, `always_allow` permission policy, unrestricted networking). The post notes that the documentation lacks dedicated security/hardening guidance (likely a beta-stage gap) and provides hardening recommendations including disabling the default toolset, using limited networking, storing all credentials in vaults, and monitoring session events. The post also notes that runtime safety classifiers (like those documented in Claude Code) could not be confirmed for Managed Agents. 

https://pluto.security/blog/inside-claude-managed-agents

Comments

Post a Comment

Popular posts from this blog

Prompt Engineering Demands Rigorous Evaluation

In the blog post “Prompt Engineering Requires Evaluation” on the Shostack + Associates website, the author argues that treating prompts for large language models (LLMs) merely as creative artefacts is insufficient. Engineering prompts properly demands structured evaluation frameworks — what the AI community calls “evals” — to test which prompt versions work better, with which models, and under which conditions. The post highlights that simply assuming a prompt is “good enough” creates risks when LLMs are integrated into production systems (e.g., for threat modeling). It advocates for measuring prompt performance, variation effects, and tool-chain dependencies (model, context, ancillary materials).  Ultimately the message is: prompt engineering should borrow disciplined practices from software engineering (versioning, testing, benchmarking) rather than relying on informal experimentation.  https://shostack.org/blog/prompt-enignieering-requires-evaluation/
Read more

SecObserve: Simplified Vulnerability and License Management for CI/CD Pipelines

 SecObserve is an open-source tool for managing vulnerabilities and licenses in software development and cloud environments. It integrates various vulnerability scanners into CI/CD pipelines using GitLab CI templates and GitHub Actions for streamlined setup. It offers a centralized dashboard for assessing and reporting vulnerabilities, with tools for filtering, sorting, and evaluating results. SecObserve supports automation and manual assessments to focus on resolving critical issues.  https://github.com/MaibornWolff/SecObserve/tree/dev
Read more

Secure Vibe Coding Guide: Best Practices for Writing Secure Code

Ken Huang's "Secure Vibe Coding Guide" emphasizes the importance of integrating security into the software development lifecycle. The guide provides best practices for writing secure code, including input validation, proper authentication mechanisms, and secure data storage techniques. It also highlights the necessity of regular code reviews and staying updated with the latest security vulnerabilities and patches. By following these guidelines, developers can create applications that are resilient against common security threats and contribute to a safer digital environment.  https://kenhuangus.substack.com/p/secure-vibe-coding-guide
Read more