> ## Content Index
> Fetch the complete content index at: https://www.agentsecbrief.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# What is a persistent approval?
- URL: https://www.agentsecbrief.com/what-is-a-persistent-approval/
- Published: 2026-11-13T14:47:43.000Z
- Updated: 2026-11-13T14:47:43.000Z
- Description: A persistent approval is a saved "don't ask again" permission that lets an AI agent repeat an action without asking the user each time.
- Author: Angela Milash
- Tags: Architecture & Harnesses, Data Exfiltration, Glossary

## What is a persistent approval?

A persistent approval is a saved "don't ask again" permission that lets an AI agent repeat an action without asking the user each time.

### Why it matters

Saved approvals reduce prompts, but they are stored somewhere, often as the exact command approved. If that command held a password or token, the approval file becomes a secret store that can be committed, shared or stolen.

### Example

Apiiro research reported by [The Hacker News](https://thehackernews.com/2026/11/claude-codes-dont-ask-again-file.html?ref=agentsecbrief.com) on 7 Nov 2026 found Claude Code's .claude/settings.local.json files in public repositories, 1,122 of them holding secrets such as database passwords and cloud keys.

### How to defend against it

- Add agent settings folders to .gitignore and secret scanning. See the [checklist](https://www.agentsecbrief.com/checklist/).
- Search git history for agent settings files and rotate anything found.
- Approve command patterns without secrets; pass credentials through a broker or environment.

### Related terms

Approval fatigue, environment-variable secrets, credential broker

### Sources

- [https://thehackernews.com/2026/11/claude-codes-dont-ask-again-file.html](https://thehackernews.com/2026/11/claude-codes-dont-ask-again-file.html?ref=agentsecbrief.com)

*From AgentSec Brief, the daily brief on securing agentic AI. Subscribe free at https://www.agentsecbrief.com*