Custom AI Detection Rules
Add a practical match for an internal or niche CLI.
Availability
Partially implemented in stable. Read the limits below before relying on this workflow.
Reviewed against stable v0.5.1. Rust candidate additions are labeled separately. Release status.
The pain this solves
- An internal CLI may not have a built-in command or output signature.
- A custom name and color can identify a recurring wrapper without claiming a full provider integration.
What ships with it
- Rules store a pattern, display name, and color in settings.
- Command lines are matched by substring, and custom patterns are included in output matching.
- A match can label a session and start AI logging; the rule does not add provider-specific proxy or usage support.
Questions this answers
- How do I add a custom AI tool to my terminal detection?
- Can I detect private or internal AI CLI tools in Chau7?
- Custom process detection rules for terminal AI monitoring
- Add new AI agent to Chau7 detection list
How it works
A custom rule stores a pattern, display name, and color in Chau7 settings. The rule editor is intended for command or output patterns, not for defining a process executable identity or a complete integration.
For submitted commands, Chau7 checks whether the lowercased command line contains the rule pattern. For terminal output, custom patterns join the built-in output-pattern matcher. A matching command can assign the custom display name and start AI logging.
Built-in detection is checked before the custom command-rule loop. Custom rules therefore extend the supported matching cases, but they do not promise a particular priority for overlapping output patterns, provider API analytics, or every behavior available to a built-in integration.
Why it matters
A team’s internal wrapper or a less common CLI may not have a built-in command signature. A custom pattern can give matching commands or output a display name; matching command text can also drive the tab color.
Frequently asked questions
What can a custom rule contain?
A pattern, display name, and color. Command matching uses substring checks, and output patterns are also included in the custom output matcher.
Does a custom rule add full support for an AI provider?
No. It can label a matching session and start AI logging, but it does not add a provider proxy adapter, usage parser, or every built-in integration behavior.
Can I share or export custom rules?
The rules are stored in Chau7 settings. The stable UI provides add and remove controls; the audited implementation does not provide a dedicated export or team-sharing workflow.