Answer: Install the runtime globally with `npm install -g @anthropic-ai/sandbox-runtime`, configure it in `~/.srt-settings.json` (or a file passed with `--settings`), then launch Claude Code through the runtime from your project directory with `srt claude`. On Linux and WSL2 you also need `bubblewrap`, `socat`, and `ripgrep`; on macOS no additional packages are needed. Status: Confirmed Entry: tool/claude-code/permissions-and-security/set-up-sandbox-runtime Revision: 1 Digest: sha256:ae9ae6aad74d8610 (changes whenever the content changes) Timeline: Created: 2026-11-13T03:05:32Z Last changed: 2026-11-13T03:05:32Z Last verified: 2026-11-20T12:24:34Z Next check: 2026-11-21T12:24:34Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of install command | npm install -g @anthropic-ai/sandbox-runtime | [1] | 2026-11-20 launch command | srt claude | [1] | 2026-11-20 default settings file | ~/.srt-settings.json | [1] | 2026-11-20 Linux/WSL2 packages | bubblewrap | [1] | 2026-11-20 Linux/WSL2 temp directory | /tmp | [1] | 2026-11-20 macOS temp directory | /private/tmp | [1] | 2026-11-20 Sources: [1] Anthropic. "Choose a sandbox environment". https://code.claude.com/docs/en/sandbox-environments.md Role: primary. Accessed: 2026-11-20T12:24:34Z Quote: "npm install -g @anthropic-ai/sandbox-runtime" (Supports: install command) Quote: "srt claude" (Supports: launch command) Quote: "Put your configuration in `~/.srt-settings.json`, or in a file you pass with `--settings`." (Supports: default settings file) Quote: "the runtime relies on the same `bubblewrap` and `socat` packages as the built-in sandbox, plus `ripgrep`" (Supports: Linux/WSL2 packages) Quote: "**Linux and WSL2**: `/tmp`" (Supports: Linux/WSL2 temp directory) Quote: "**macOS**: `/private/tmp`" (Supports: macOS temp directory) Published: 2026-11-20T03:41:01Z Snapshot SHA-256: f3c482d1ba678403dfa26c1595e08c9a380a880d27dcb50e583d0f201e579aff (captured 2026-11-20T12:24:32Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. Cross-check: not-available at 2026-11-20T12:24:34Z. No independent second source is configured for this entry Last successful check: 2026-11-20T12:24:34Z Last attempt: 2026-11-20T12:24:34Z Notes and limitations: Extracted automatically from a single official source (section "Set up and launch the runtime"). Related: https://latestn.com/tool/claude-code/permissions-and-security/slack-access-control How do I control who can access Claude Code in Slack? https://latestn.com/tool/claude-code/permissions-and-security/stop-computer-use How do I stop Claude from controlling my screen during computer use? https://latestn.com/tool/claude-code/permissions-and-security/subagent-tool-restrictions How do I restrict which tools a subagent can use in the Claude Agent SDK? https://latestn.com/tool/claude-code/permissions-and-security/security-best-practices-sensitive-code What security best practices should I follow when using Claude Code with sensitive code? https://latestn.com/tool/claude-code/permissions-and-security/security-plugin-file-edit-checks What does the Claude Code security-guidance plugin check on each file edit? https://latestn.com/tool/claude-code/permissions-and-security/set-permission-mode How do I set the permission mode in Claude Code? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T03:05:32Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/claude-code/permissions-and-security/set-up-sandbox-runtime/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/claude-code/permissions-and-security/set-up-sandbox-runtime POST: https://latestn.com/dispute/tool/claude-code/permissions-and-security/set-up-sandbox-runtime with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/claude-code/permissions-and-security/set-up-sandbox-runtime?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How do I set up and launch the Claude Code sandbox runtime? Revision 1, last verified 2026-11-20T12:24:34Z. https://latestn.com/tool/claude-code/permissions-and-security/set-up-sandbox-runtime (accessed 2026-11-21T09:58:59Z). @misc{LatestN_tool_claude_code_permissions_and_security_set_up_sandbox_runtime_r1, title = {How do I set up and launch the Claude Code sandbox runtime?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/claude-code/permissions-and-security/set-up-sandbox-runtime}}, note = {Revision 1; digest sha256:ae9ae6aad74d8610; last verified 2026-11-20T12:24:34Z; accessed 2026-11-21T09:58:59Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/claude-code/permissions-and-security/set-up-sandbox-runtime