mirror of
https://github.com/NickCrew/Claude-Cortex.git
synced 2026-09-14 20:17:11 +08:00
35ebb2abe3
Remove cortex-workflow references since the command is not implemented and workflow orchestration is currently handled through composition of existing commands (agent, skills, review, plan, export, ai). - Remove workflow subcommand from validation and generation - Remove cortex-workflow.1 from manpage bundle - Delete placeholder cortex-workflow.1 manpage file This keeps the build clean and only bundles manpages for implemented commands (cortex.1 and cortex-tui.1).
3.9 KiB
3.9 KiB
layout, title, parent, nav_order
| layout | title | parent | nav_order |
|---|---|---|---|
| default | Configuration Reference | Reference | 1 |
Configuration Reference
This guide covers the configuration files and resolution rules used by the current Cortex CLI.
Resolution model
Cortex resolves two different roots:
- Asset root (CORTEX_ROOT, or
--cortex-root)
Used for bundled assets and watch defaults. - Claude directory (from
--scope/ CORTEX_SCOPE)
Used for user/project state under.claude/.
--scope supports auto, project, and global.
project: nearest.claude/in the current directory tree (or creates one in cwd)global:~/.claude/auto(default): nearest.claude/, else~/.claude/
Quick reference
| File | Location | Purpose |
|---|---|---|
cortex-config.json |
<CORTEX_ROOT>/cortex-config.json |
Watch-mode defaults consumed by cortex ai watch |
recommendation-rules.json |
<resolved .claude>/skills/recommendation-rules.json |
File-pattern-based skill recommendations |
skill-rules.json |
<CORTEX_ROOT>/skills/skill-rules.json (watch fallback: ~/.claude/skills/skill-rules.json) |
Keyword-based skill suggestions |
settings.json |
<resolved .claude>/settings.json |
Claude settings used by hooks/statusline configuration |
.onboarding-state.json |
<resolved .claude>/.onboarding-state.json |
Optional onboarding-state schema target |
memory-config.json |
<resolved .claude>/memory-config.json |
Optional memory-config schema target |
cortex-config.json
cortex-config.json is currently used by watch mode (cortex ai watch) for defaults.
Supported keys (watch block)
- watch.directories (or watch.dirs): list of directories
- watch.auto_activate: boolean
- watch.threshold: float between
0.0and1.0 - watch.interval: polling interval in seconds (
> 0)
Example
{
"watch": {
"directories": ["~/Developer/my-project"],
"auto_activate": true,
"threshold": 0.75,
"interval": 2.0
}
}
Usage
# Uses defaults from <CORTEX_ROOT>/cortex-config.json when present
cortex ai watch
# Override defaults at runtime
cortex ai watch --dir . --threshold 0.8 --interval 1.5
recommendation-rules.json
Schema: schemas/recommendation-rules.schema.json
Minimal example
{
"version": "2026-02-22",
"rules": [
{
"trigger": {
"file_patterns": ["**/auth/**", "**/security/**"]
},
"recommend": [
{
"skill": "secure-coding-practices",
"confidence": 0.9,
"reason": "Security-sensitive files changed"
}
]
}
]
}
skill-rules.json
Schema: schemas/skill-rules.schema.json
Minimal example
{
"version": "2026-02-22",
"rules": [
{
"name": "debugging",
"command": "/ctx:skill systematic-debugging",
"description": "Recommend structured debugging when users report failures.",
"keywords": ["debug", "failing", "error"]
}
]
}
Optional schemas
These schemas are present in schemas/ and can be used for validation/autocomplete in editors:
Practical commands
# Check which scope/root you are using
cortex --scope project status
cortex --scope global status
# Point CLI at a specific Cortex asset root
cortex --cortex-root /path/to/claude-cortex status
# Inspect command-specific options
cortex ai watch --help