Agent Claude Code
Local Devstral agent for LiteCLI-powered tasks. Use when the /devstral skill spawns a local agent.
Kuroryuu - AI Agent Orchestration Platform with Desktop App, MCP Integration, and Docker Support
This repository also configures its own agents. See what kuroryuu-public tells them →
Agent Claude Code
Local Devstral agent for LiteCLI-powered tasks. Use when the /devstral skill spawns a local agent.
Agent Claude Code
Implements ONE task for /k-start-worker-loop. Verifies work. Does NOT commit.
Agent Claude Code
Kuroryuu deep explorer - thorough read-only research and codebase analysis with Opus reasoning.
Agent Claude Code
Kuroryuu explorer agent - read-only research and codebase analysis, fast exploration.
Agent Claude Code
Generates new Claude Code sub-agent configuration files from descriptions. Use proactively when the user asks to create a new sub-agent.
Agent Claude Code
Challenges reasoning quality of plans, decisions, and approaches through 6 meta-cognitive lenses. Use when reviewing plans, PRDs, architectural decisions, or any significant reasoning output. Read-only — analyzes documents and produces structured reports. Spawn before committing to architecture decisions, when a plan…
Agent Claude Code
Max-Parallel builder that executes ONE task with self-validation. Focused context window for code writing, file creation, implementation work. Use with /max-subagents-parallel workflows.
Agent Claude Code
Read-only validator for /max-subagents-parallel workflows. Verifies task completion against acceptance criteria. Cannot modify files - inspects and reports only.
Agent Claude Code
Thinks through Marc Andreessen's cognitive architecture — identifying technological discontinuities, holding strong opinions loosely for maximum learning velocity, and synthesizing cross-domain patterns. Use for market timing, technology adoption curves, platform shifts, and contrarian bets.
Agent Claude Code
Reasons through Marcus Aurelius's cognitive architecture — dichotomy of control, premeditatio malorum, the obstacle is the way, and daily journaled reflection. Forces separation of what is within your power from what is not. Use for decision-making under pressure, resilience, perspective, and daily reflection practice.
Agent Claude Code
Reasons through Jeff Bezos's cognitive architecture — working backwards from the customer, PR/FAQ forcing functions for clarity, two-way vs one-way door decision speed, and Day 1 thinking to resist entropy. Forces customer announcement before any feature discussion. Use for customer-obsessed design, decision speed…
Agent Claude Code
Reasons through Carmack's cognitive architecture — constraint-first engineering, mathematical shortcuts, paper-to-production pipeline, anti-abstraction discipline. Forces identification of the actual bottleneck before any optimization. Use for performance work, systems architecture, code review, or technical…
Agent Claude Code
Reasons through Leonardo da Vinci's cognitive architecture — saper vedere (knowing how to see), mechanism-hunting across domains, observation before theory, cross-domain transfer via structural analogy. Forces direct observation and mechanism identification before any solution. Use for cross-disciplinary synthesis…
Agent Claude Code
Reasons through Walt Disney's cognitive architecture — the Dreamer/Realist/Critic triad, storyboarding, Plussing, and Blue Sky ideation. Forces full separation of creative phases before convergence. Use for experience design, creative strategy, storyboarding, and iterative improvement.
Agent Claude Code
Reasons through Feynman's cognitive architecture — first-principles rebuilding, the freshman test, twelve problems cross-referencing, play as cognitive strategy. Forces explanation from intuition through formalism. Use for debugging, learning new domains, explaining complex concepts, or detecting cargo cult thinking.
Agent Claude Code
Reasons through Erich Gamma's cognitive architecture — refactor to patterns not from patterns, composition over inheritance, the Rule of Three, sufficiency over completeness, pattern removal when simpler wins. Forces feeling the design pain before applying any pattern. Use for code architecture, API design, framework…
Agent Claude Code
Reasons through Bill Gates's cognitive architecture — platform thinking, systematic deep-reading, multi-source learning, decomposition into atomic components, and time dynamics mapping. Forces proper modeling before betting and keystone layer identification. Use for platform strategy, ecosystem design, deep analysis…
Agent Claude Code
Thinks through Seth Godin's cognitive architecture — smallest viable audience, worldview-first positioning, and permission over interruption. Use for marketing strategy, audience building, positioning, and "would anyone remark about this?".
Agent Claude Code
Thinks through Paul Graham's cognitive architecture — pattern observation, unscaled experimentation, and essay-driven clarity. Use for startup strategy, product-market fit, writing to think, and founder evaluation.
Agent Claude Code
Reasons through Richard Helm's cognitive architecture — engineering handbook pattern lookup, behavioral contracts between collaborating objects, tradeoffs over prescriptions, runtime composition over static structure. Forces explicit tradeoff analysis before any design commitment. Use for architectural evaluation…
Agent Claude Code
Reasons through Steve Jobs's cognitive architecture — taste at the intersection of technology and liberal arts, radical simplification, reality distortion as constraint reframing, and shipping what people don't know they need. Forces aesthetic-first evaluation before any feature discussion. Use for product vision, UX…
Agent Claude Code
Reasons through Ralph Johnson's cognitive architecture — concrete examples before abstraction, frameworks as components plus patterns, architecture as shared understanding, white-box to black-box evolution. Forces three concrete cases before any generalization. Use for framework design, library architecture…
Agent Claude Code
Reasons through Linus Torvalds's cognitive architecture — good taste in code as structural elegance, working code as the only valid argument, pragmatic empiricism over theory, and minimal abstraction. Forces "show me the code" before any discussion. Use for code review, architecture taste assessment, BS detection, or…
Agent Claude Code
Reasons through Ada Lovelace's cognitive architecture — identifying operational structure beneath specific implementations, abstracting patterns to their general class, poetical science (imagination fused with mathematical rigor). Forces the question "what else has this structure?" Use for technology visioning, system…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: