Overview
- Stop API keys, tokens, and private keys from leaking to ChatGPT, Claude, or Cursor by scanning files and staged git changes before they ever reach an AI tool with Offsend's local-first check command.
- See exactly which sensitive files and folders are visible to AI tools like Cursor or Claude before you paste or share context using the read-only show command that categorizes exposed paths by data type.
- Block commits containing privileged information automatically by installing a pre-commit git hook that screens every staged change for secrets and rejects the commit if anything sensitive is found.
- Enforce AI data boundaries per repository without manual ignore file management by running the prepare command to generate missing .cursorignore, .claudeignore, and .aiexclude files in seconds.
- Tailor secret detection to your specific tech stack across every project by committing a .offsend.yml configuration file that tunes detectors, exclusions, and custom dictionaries for each repository.
- Run security checks in your CI pipeline without sending code to any server by integrating the local-first CLI into automated workflows—no account, no upload, no cloud dependency.
- Install and start protecting your codebase in seconds on macOS with a single brew command, no signup required, using the free open-source CLI that operates entirely on your machine.
Pros & Cons
Pros
- Free, open-source tool
- Local-first tool
- Scans files for secrets
- Reviews sensitive file paths
- Terminal compatible
- Can operate with git hooks
- Works with Continuous Integration
- No server upload requirement
- Supports per repository configuration
- Allows custom detectors and exclusions
- Tightens risk management
- Supports macOS
- Local pre-commit hook feature
- Advanced secrets management
- Automatic generation of ignore rules
- Offers private key security
- API key protection
- Token security mechanism
- CLI-tool command options
- Secure data handling
- Compliance-optimized tool
- Terminal demo for commands
- Prepares projects ignore files
- Configurable and automation-oriented
- No account creation needed
- Homebrew-compatible installation
- Provides a .offsend.yml file
- Lists exposed sensitive files
- Blocks commits containing secrets
- On-premise safety management
- Secure coding support
- Code exposure audit functionality
- Protects sensitive project context
- API-agnostic; works with any
- Supports custom dictionaries
- Excludes specified file types
- Command 'offsend show' for exposure list
- Command 'offsend prepare' for file generation
- Command 'offsend hook install' for pre-commit hook
- Command 'offsend check' for scanning files
- Handles data protection
- Optimized for software development
- Enforces advanced infosec standards
- Improved DevOps security
- Provides continuous integration compatibility
- Enhances the confidentiality of projects
Cons
- Only available on macOS
- Limited to local scanning
- No support for Windows/Linux
- No cloud-based scanning option
- No web-based interface
- No multi-user support
- Configuration limited to .offsend.yml
- No built-in alert system
- No integration with issue trackers
- No support for non-git projects
Reviews
Rate this tool
Loading reviews...
❓ Frequently Asked Questions
Offsend CLI is a free, local-first command-line tool designed to protect sensitive data in AI contexts. It works by scanning files and staged changes, auditing what is visible to AI tools, operating from terminals, git hooks, or CI environments, and blocking commits containing sensitive or privileged information.
No, Offsend CLI operates without requiring any user account.
Yes, Offsend CLI can be used directly from terminals, git hooks, and continuous integration (CI) environments.
Offsend CLI protects sensitive data in AI contexts by scanning files and staged changes for API keys, tokens, private keys, and similar secrets. It audits what is visible and accessible to AI tools and blocks commits containing privileged information.
The command-line interface of Offsend CLI is equipped with four commands: 'offsend check', 'offsend show', 'offsend prepare', and 'offsend hook install'.
Yes, Offsend CLI allows you to specify detectors, exclusions, and custom dictionaries based on your needs.
'Offsend check' command works by scanning files, folders, or staged git changes for API keys, tokens, private keys, and similar secrets.
'Offsend show' command lists the sensitive files that are not covered by ignore rules.
'Offsend prepare' is used to create new AI ignore files where required, which helps to enforce boundaries for AI tools.
'Offsend hook install' performs the function of installing a pre-commit hook that runs checks against staged changes and blocks any commits containing sensitive information.
Checks per repository can be tuned using a .offsend.yml file in Offsend CLI. This file allows to specify detectors, exclusions, and custom dictionaries, tailoring the checks to the requirements of the specific repository.
The desktop app version of Offsend CLI offers additional features such as Safe Paste, one-click fixes, and local AI detection.
Yes, Offsend CLI is an open-source tool.
Offsend CLI can be installed on macOS.
No, there is no server-side scanning required for Offsend CLI as it is a local-first tool.
Yes, Offsend CLI can be used to block commits that contain sensitive or privileged information.
Data protection is ensured by Offsend CLI by preventing privileged data from being visible to AI tools. It conducts checks to find sensitive data (like API keys, tokens, private keys, etc.) and blocks commits that contain such information.
The .offsend.yml file in Offsend CLI serves to configure and automate checks. It allows tuning of detectors, exclusions, and custom dictionaries. CLI flags override these config values when provided explicitly.
Offsend CLI can be installed and ready to use in seconds, aiding in rapid implementation of data security measures.
Yes, Offsend CLI can be used on git hooks.
Offsend CLI protects sensitive project context by scanning files, folders, and staged git changes for secrets such as API keys, tokens, private keys, credentials, and other sensitive information. It shows what may be exposed to AI tools and helps generate AI ignore files. Moreover, it can be used from the terminal, git hooks, or CI without sending code to a server.
Yes, Offsend CLI can indeed be used in Continuous Integration (CI) environments. It is designed to run locally without requiring an account or moving data to a cloud server, making it suitable for automation and configuration in CI pipelines.
Offsend CLI scans for secrets such as API keys, tokens, private keys, credentials, and other categories of secure data. It identifies these secrets before they get exposed to AI workflows, thus preventing potential security risks.
You can use the Offsend CLI from the terminal by utilizing several commands. For example, 'offsend check -staged' scans your staged changes, files, and folders for sensitive data, 'offsend show' lists the exposed sensitive paths that can be accessed by AI tools, and 'offsend prepare' crafts AI ignore files for the project.
No, Offsend CLI operates in a 'local-first' mode, which means it carries out all operations on your local machine and does not send any file content or code to a server. This provides an added layer of protection to your sensitive data.
You can use the 'offsend check' command to scan files, folders or staged git changes for secrets such as API keys, tokens, private keys, and similar sensitive information before they leave your Mac. This command can be run in the terminal.
The 'offsend show' command lists the sensitive files that would be sent to the AI too.l because no ignore rule covers them. The results are categorised by data type and are read-only.
You can generate AI ignore files using the command 'offsend prepare'. Running this command in the terminal will create the AI ignore files that your project is missing so that AI education tools respect your data boundaries.
The 'offsend hook install' command installs a pre-commit hook that runs the same checks against staged changes and blocks commits containing sensitive data. This provides an added layer of protection that pre-screens staged changes before the commit occurs.
Yes, the Offsend CLI can be configured for each repository by committing a .offsend.yml file into your repository. This allows developers to adjust detectors, exclusions, and custom dictionaries according to their stack.
The '.offsend.yml' file allows developers to tune detectors, exclusions, and customise dictionaries for each repository. This feature allows checks to be tuned and run in CI. CLI flags can override config values when explicitly provided.
Yes, Offsend CLI supports macOS. It is designed to run locally and can be installed easily on macOS systems.
You can use the command 'brew install -cask offsend/tap/offsend-cli' to install Offsend CLI on your local macOS machine.
Yes, Offsend CLI is a free command-line tool. It is also open-source, allowing anyone to utilise, modify, and distribute the software.
Offsend CLI contributes to AI safety by scanning and alerting developers of sensitive information in files, folders, and staged git changes that might be inadvertently exposed to AI tools. It generates AI ignore files to keep confidential data out of AI's reach, and it operates from the terminal without requiring an account or sending code to a server. This provides a safeguard against accidental data exposure.
A local-first tool like Offsend CLI has the advantage of conducting all operations on the user's local machine without the need to upload file contents to a server. This ensures sensitive information remains within the premises, thus reducing the risk of data exposure.
Yes, Offsend CLI can be integrated into your git hooks. By using the command 'offsend hook install', a pre-commit hook that screens staged changes and prevents committing of secure data can be installed.
Yes, Offsend CLI is an open-source software. Its source code is freely available for modification and distribution.
Offsend CLI prevents sensitive data from being exposed to AI tools by scanning files, folders, and staged git changes for secrets such as API keys, tokens, private keys, credentials, and other sensitive data. It alerts developers of any potential exposure before the data reaches the AI tools. Furthermore, it aids in generating AI ignore files that ensure confidential data remains outside the AI's reading context. Importantly, all operations are prosecuted on the local machine without the need for server-side scanning or accounts.
Pricing
Pricing model
Free
Paid options from
Free
