Configure models, tools, and options for kenken workflow
Configures models, tools, and options for the kenken workflow via an interactive wizard or command arguments.
npx claudepluginhub kenkenmain/ken-cc-pluginsThis skill inherits all available tools. When active, it can use any tool Claude has access to.
For Claude: This skill manages kenken configuration. Parse arguments and execute accordingly.
| Argument | Action |
|---|---|
| (none) | Interactive configuration wizard |
--show | Display current merged config |
--reset | Reset config to defaults |
Display current configuration by:
~/.claude/kenken-config.json (if exists).claude/kenken-config.json (if exists)kenken Configuration
Block on severity: low
PLAN Stage:
brainstorm: model=inherit, parallel=true
writePlan: model=inherit, parallel=true
planReview: tool=mcp__codex-high__codex
IMPLEMENT Stage:
implementation: model=inherit, implementer=claude, bugFixer=codex-high, enforceLogging=true
simplify: model=inherit, bugFixer=codex-high
implementReview: tool=mcp__codex-high__codex, bugFixer=codex-high
TEST Stage: [disabled]
instructions: (not configured)
commands: lint=make lint, test=make test, coverage=make coverage
coverageThreshold: 80%
testPlan: model=inherit
testImplementation: model=inherit
testReview: tool=mcp__codex-high__codex
FINAL Stage:
codexFinal: tool=mcp__codex-xhigh__codex (fixed)
suggestExtensions: enabled=true, maxSuggestions=3
Config files:
Global: ~/.claude/kenken-config.json [found/not found]
Project: .claude/kenken-config.json [found/not found]
Ask user which config to reset:
Use AskUserQuestion:
Delete selected file(s)
Confirm: "Reset complete. Using defaults."
Display current values (as in --show).
Use AskUserQuestion with multiSelect:
For PLAN Stage:
Ask for each phase:
Common model options: inherit, sonnet, opus, haiku, opus-4.5, sonnet-4, sonnet-3.5
For IMPLEMENT Stage:
For TEST Stage:
For FINAL Stage:
Ask: "Where to save?"
{
"version": 1,
"blockOnSeverity": "low",
"stages": {
"plan": {
"brainstorm": { "model": "inherit", "parallel": true },
"writePlan": { "model": "inherit", "parallel": true },
"planReview": { "tool": "mcp__codex-high__codex", "maxRetries": 3 }
},
"implement": {
"implementation": {
"model": "inherit",
"implementer": "claude",
"bugFixer": "codex-high",
"enforceLogging": true
},
"simplify": { "model": "inherit", "bugFixer": "codex-high" },
"implementReview": {
"tool": "mcp__codex-high__codex",
"bugFixer": "codex-high",
"maxRetries": 3
}
},
"test": {
"enabled": false,
"instructions": "",
"commands": {
"lint": "make lint",
"test": "make test",
"coverage": "make coverage"
},
"coverageThreshold": 80,
"coverageFormat": "auto",
"testPlan": { "model": "inherit" },
"testImplementation": { "model": "inherit", "bugFixer": "codex-high" },
"runTests": { "timeout": 300 },
"testReview": {
"tool": "mcp__codex-high__codex",
"bugFixer": "codex-high",
"maxRetries": 3
}
},
"final": {
"codexFinal": {
"tool": "mcp__codex-xhigh__codex",
"bugFixer": "codex-high"
},
"suggestExtensions": { "enabled": true, "maxSuggestions": 3 }
}
},
"git": {
"branchFormat": "{type}/{slug}",
"defaultType": "feat",
"mainBranch": "auto"
},
"logging": {
"directory": ".agents/logs",
"retainDays": 7,
"extractErrors": true
}
}
| Setting | Default |
|---|---|
| blockOnSeverity | low |
| All models | inherit |
| Implementer | claude |
| Bug fixer | codex-high |
| Review tools | mcp__codex-high__codex |
| Final tool | mcp__codex-xhigh__codex |
| Test stage | disabled |
| Coverage threshold | 80% |
| Max retries | 3 |
| Extensions | enabled |
| Test instructions | (must be provided) |
| Branch format | {type}/{slug} |
| Default type | feat |
| Main branch | auto (detect) |
| Setting | Valid Values |
|---|---|
| blockOnSeverity | high, medium, low (blocks on specified level and above) |
| model | inherit, or any valid model name (see Model Names below) |
| implementer | claude, codex-high, codex-xhigh (who writes initial code) |
| bugFixer | claude, codex-high, codex-xhigh (who fixes issues found by reviews) |
| tool (review) | mcp__codex-high__codex, mcp__codex-xhigh__codex, claude-review (see note) |
| tool (final) | mcp__codex-xhigh__codex only (fixed) |
| threshold | 0-100 |
| maxRetries | 1-10 |
| timeout | 60-3600 (seconds) |
| maxSuggestions | 1-5 |
| instructions | non-empty string (required when test.enabled=true) |
| commands.* | valid shell command string |
| coverageFormat | auto, lcov, cobertura, json |
| branchFormat | string with placeholders: {type}, {slug}, {date}, {user} |
| defaultType | feat, fix, chore, refactor, docs, test |
| mainBranch | auto, main, master, or custom branch name |
Note on claude-review: This option uses the superpowers:requesting-code-review skill instead of Codex MCP. It's available as a fallback when Codex MCP is not configured, or for users who prefer Claude-native reviews. No additional dependencies required.
The model field accepts several formats:
sonnet - Claude Sonnet (latest)opus - Claude Opus (latest)haiku - Claude Haiku (latest)opus-4.5 - Claude Opus 4.5opus-4 - Claude Opus 4sonnet-4 - Claude Sonnet 4sonnet-3.5 - Claude Sonnet 3.5haiku-3.5 - Claude Haiku 3.5For maximum control, use the full model identifier:
claude-opus-4-5-20251101claude-sonnet-4-20250514claude-sonnet-3-5-20241022inherit - Use the current session's model (recommended default)Activates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.
Search, retrieve, and install Agent Skills from the prompts.chat registry using MCP tools. Use when the user asks to find skills, browse skill catalogs, install a skill for Claude, or extend Claude's capabilities with reusable AI agent components.
This skill should be used when the user wants to "create a skill", "add a skill to plugin", "write a new skill", "improve skill description", "organize skill content", or needs guidance on skill structure, progressive disclosure, or skill development best practices for Claude Code plugins.