Answer: A `tool.call` hook can `await` before it calls `next` or returns, keeping the tool call pending, and it can call `$.ui.ask` to show the question with numbered options to the user. Status: Confirmed Entry: tool/claude-code/permissions-and-security/hold-tool-call-for-user Revision: 3 Digest: sha256:688f5fcd4fa4523b (changes whenever the content changes) Timeline: Created: 2026-11-13T03:05:13Z Last changed: 2026-11-18T12:18:53Z (timeout behavior: — → Claude Code skips a hook that times out, so the held command would run.) Last verified: 2026-11-20T12:23:39Z Next check: 2026-11-21T12:23:39Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of mechanism | await | [1] | 2026-11-20 ask method | $.ui.ask | [1] | 2026-11-20 example options | ['Run it', 'Refuse'] | [1] | 2026-11-20 timeout behavior | Claude Code skips a hook that times out, so the held command would run. | [1] | 2026-11-20 Sources: [1] Anthropic. "React to events with a mod". https://code.claude.com/docs/en/plugins/mods/events.md Role: primary. Accessed: 2026-11-20T12:23:39Z Quote: "A `tool.call` hook can `await` before it calls `next` or returns, and the tool call stays pending until then." (Supports: mechanism) Quote: "To put the question to the user, call `$.ui.ask`." (Supports: ask method) Quote: "answer = await $.ui.ask('Run this command? ' + e.command, ['Run it', 'Refuse'])" (Supports: example options) Quote: "Claude Code skips a hook that times out, so the held command would run." (Supports: timeout behavior) Published: 2026-11-20T03:41:00Z Snapshot SHA-256: b1a98246633f1352cfea4789628969e972a9fdb0b126114925ff98a69a589326 (captured 2026-11-20T12:23:39Z) 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 2026-11-20T12:23:39Z. No independent second source is configured for this entry Last successful check: 2026-11-20T12:23:39Z Last attempt: 2026-11-20T12:23:39Z Notes and limitations: Extracted automatically from a single official source (section "Hold a tool call until the user decides"). Related: https://latestn.com/tool/claude-code/permissions-and-security/iam-configuration What IAM permissions does Claude Code need on Google Cloud's Agent Platform? https://latestn.com/tool/claude-code/permissions-and-security/iam-permissions What IAM permissions are required for Claude Code on Amazon Bedrock? https://latestn.com/tool/claude-code/permissions-and-security/isolation-technologies What isolation technologies can I use to securely deploy Claude Code agents? https://latestn.com/tool/claude-code/permissions-and-security/filesystem-configuration How do I configure filesystem access and writable locations for a Claude Code deployment? https://latestn.com/tool/claude-code/permissions-and-security/foundry-azure-credentials How do I authenticate Claude Code with Microsoft Foundry using an API key, Microsoft Entra ID, or a bearer token? https://latestn.com/tool/claude-code/permissions-and-security/gate-inbound-messages How do I gate inbound messages in a Claude Code channel to only trusted senders? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-18T12:18:53Z | changed | timeout behavior: — → Claude Code skips a hook that times out, so the held command would run. | [1] 2026-11-18T12:18:53Z | changed | deny response: { deny: 'The user declined this command. Ask before trying a different approach.' } → — | [1] 2026-11-15T12:23:53Z | changed | deny response: — → { deny: 'The user declined this command. Ask before trying a different approach.' } | [1] 2026-11-15T12:23:53Z | changed | example options: — → ['Run it', 'Refuse'] | [1] 2026-11-15T12:23:53Z | changed | refuse behavior: deny → — | [1] 2026-11-15T12:23:53Z | changed | example labels: ['Run it', 'Refuse'] → — | [1] 2026-11-13T03:05:13Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/claude-code/permissions-and-security/hold-tool-call-for-user/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/hold-tool-call-for-user POST: https://latestn.com/dispute/tool/claude-code/permissions-and-security/hold-tool-call-for-user with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/claude-code/permissions-and-security/hold-tool-call-for-user?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How do I make a Claude Code mod hook hold a tool call until the user decides? Revision 3, last verified 2026-11-20T12:23:39Z. https://latestn.com/tool/claude-code/permissions-and-security/hold-tool-call-for-user (accessed 2026-11-21T10:03:11Z). @misc{LatestN_tool_claude_code_permissions_and_security_hold_tool_call_for_user_r3, title = {How do I make a Claude Code mod hook hold a tool call until the user decides?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/claude-code/permissions-and-security/hold-tool-call-for-user}}, note = {Revision 3; digest sha256:688f5fcd4fa4523b; last verified 2026-11-20T12:23:39Z; accessed 2026-11-21T10:03:11Z} } 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/hold-tool-call-for-user