TL;DR
Get privacy and security gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Apiiro reports finding passwords in 1,122 public GitHub files associated with Claude Code’s “Don’t Ask Again” setting. The source material does not establish how many credentials are active, whether they were accessed, or whether the setting itself caused the exposures.
Security company Apiiro says it found passwords in 1,122 publicly available GitHub files associated with Claude Code users selecting the tool’s “Don’t Ask Again” option. The finding points to a possible route for development secrets to enter public code repositories, but the available report does not establish how many credentials remain usable or whether anyone accessed them.
Apiiro’s report describes the files as publicly exposed and connects them to use of Claude Code’s “Don’t Ask Again” choice. The supplied report text gives the headline count, but does not include the underlying search method, the time period covered, the repositories examined, or a breakdown of the password types. The 1,122 figure should therefore be treated as Apiiro’s reported count, not as an independently verified total.
The source material also does not explain the precise sequence by which the passwords entered the files. It does not provide examples, identify affected projects or account owners, or state whether the files were committed directly by developers, generated as part of agent activity, or added through another workflow. The association with a user-selected option is part of Apiiro’s report; it does not, by itself, show that Claude Code exposed credentials automatically.
No independent confirmation, response from Anthropic, or specific remediation guidance appears in the material provided. It also does not say whether Apiiro notified affected repository owners or whether any credentials have since been removed or rotated. Those omissions limit what can be concluded about immediate exposure and the status of the reported files.
Why Public Credential Exposure Matters
Passwords committed to public repositories can be copied by anyone with access to the files. If a credential is valid and grants access to a service, an attacker could potentially use it to reach accounts, infrastructure, or data; the actual risk depends on each password’s permissions and whether it has been revoked. Apiiro’s count signals a possible exposure pattern, but does not establish that any account was compromised.
The report also focuses attention on how coding assistants and their surrounding workflows interact with sensitive information. Developers may use agents while working in repositories that contain configuration files, test data, or secrets. A setting that reduces repeated prompts could affect user decisions during that work, but the supplied evidence does not show the setting’s technical behavior or prove that it caused the files to become public. For developers, the practical concern is whether credentials were committed and remain active, not the option’s label alone.
Products Worth Considering
Apple iPad 11-inch with A16 Chip, 512GB, Wi-Fi 6, and Apple Pencil (USB-C) — Blue
Apple MacBook Air 13.6-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Sky Blue
Logitech MX Creative Console with 9 Customizable Keys, Graphite
As an affiliate, we earn on qualifying purchases.
Claude Code and the Reported Setting
Claude Code is Anthropic’s coding tool, while GitHub hosts code repositories that may be public or private. The report title links the alleged password exposure to a setting called “Don’t Ask Again,” but the source text supplied here contains no technical explanation of what that setting suppresses, when it appears, or how it stores user choices.
That distinction matters: a report that identifies files in public repositories is not the same as a finding that an AI model independently disclosed secrets. The available material documents Apiiro’s claim and the reported file count, but offers too little detail to assign responsibility among the tool, the user, repository practices, or other software involved.
“Claude Code’s “Don’t Ask Again” exposes passwords in 1,122 public GitHub files”
— Apiiro, in the report headline
Credential Status and Cause Unclear
The supplied source does not say whether the reported passwords are current, how many distinct credentials or owners are involved, or whether any were used to access accounts. It gives no date range or method for identifying the 1,122 files, and does not clarify whether the count refers to unique files, repositories, or another unit beyond the headline’s wording.
It is also unclear what role the Claude Code option played in each case and whether the files remain public. No response from Anthropic, affected developers, or GitHub is included. Without those details, the report cannot establish the cause, the scale of any successful misuse, or whether the exposures have been addressed.
Verification and Credential Remediation
Further reporting would need Apiiro’s methodology, the period covered, and details on how it linked the files to the “Don’t Ask Again” option. Responses from Anthropic and GitHub could clarify the setting’s behavior and whether either company has investigated the reported cases. The supplied material gives no date for such updates.
Repository owners whose credentials may have been exposed would need to check whether the secrets are still valid, revoke or rotate them if necessary, and remove sensitive material from public code. Deleting a file alone may not address copies already made. Whether these steps have occurred for any of the reported files is not stated.
Key Questions
What did Apiiro report?
Apiiro says it found passwords in 1,122 public GitHub files associated with Claude Code’s “Don’t Ask Again” option. The supplied source does not include independent verification of that count.
Does the report prove Claude Code leaked the passwords?
No. The source links the files to the option, but does not explain the technical cause or establish that Claude Code itself placed passwords in public repositories.
Were accounts or services compromised?
The supplied material does not say whether any passwords were active or used. It reports exposed files, not confirmed account access or a breach.
What should repository owners do if they find a password in public code?
They should treat a publicly exposed credential as potentially copied, check whether it is still valid, and revoke or rotate it if needed. The source does not say what actions affected owners have taken.
Source: apiiro.com
