caee558e79
Add /ultraresearch-local for structured research combining local codebase analysis with external knowledge via parallel agent swarms. Produces research briefs with triangulation, confidence ratings, and source quality assessment. New command: /ultraresearch-local with modes --quick, --local, --external, --fg. New agents: research-orchestrator (opus), docs-researcher, community-researcher, security-researcher, contrarian-researcher, gemini-bridge (all sonnet). New template: research-brief-template.md. Integration: --research flag in /ultraplan-local accepts pre-built research briefs (up to 3), enriches the interview and exploration phases. Planning orchestrator cross-references brief findings during synthesis. Design principle: Context Engineering — right information to right agent at right time. Research briefs are structured artifacts in the pipeline: ultraresearch → brief → ultraplan --research → plan → ultraexecute. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2.1 KiB
2.1 KiB
settings.json Schema Reference
File Locations
| Location | Precedence | Purpose |
|---|---|---|
~/.claude/settings.json |
Lowest | User defaults |
.claude/settings.json |
Medium | Project shared |
.claude/settings.local.json |
High | Project local |
| CLI arguments | Highest | Session only |
Schema
{
// Default model for the project
"model": "sonnet",
// Permission rules
"permissions": {
// Tools allowed without prompting
"allow": [
"Read",
"Write",
"Bash(npm*)",
"Bash(git*)"
],
// Tools that always require approval
"deny": [
"Bash(rm -rf*)"
]
},
// Environment variables to set
"env": {
"NODE_ENV": "development"
},
// Hooks configuration
"hooks": {
"PreToolUse": [...],
"PostToolUse": [...],
"Stop": [...]
},
// MCP server configuration (can also be in .mcp.json)
"mcpServers": {
"filesystem": {
"command": "npx",
"args": ["-y", "@anthropic/mcp-server-filesystem"],
"env": {}
}
},
// Custom agents path
"agents": "./agents",
// Plugins to load
"plugins": [
"~/plugins/my-plugin"
]
}
Key Settings
model
Default model for this project/user:
{
"model": "sonnet" // or "opus", "haiku"
}
permissions
Control tool access:
{
"permissions": {
"allow": [
"Read",
"Write",
"Bash(npm *)",
"Bash(git *)",
"Task"
],
"deny": [
"Bash(rm -rf *)",
"Bash(sudo *)"
]
}
}
Patterns support wildcards:
*matches any charactersBash(npm*)matchesnpm install,npm test, etc.
env
Environment variables:
{
"env": {
"NODE_ENV": "development",
"DEBUG": "true"
}
}
hooks
Event-driven automation:
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"command": "echo 'About to run bash'"
}
]
}
}
Merging Behavior
When multiple settings files exist:
- Objects are merged recursively
- Arrays are replaced, not merged
- Higher precedence wins for conflicts