Agent skill · research science · nexu-io
deep-think-maximum-cognitive-effort-protocol-mq8kvw92
Use this plugin when the user wants a maximum-effort reasoning workflow for a complex, high-stakes, or ambiguous task.
Why this skill is useful
Provides a structured, multi-phase reasoning workflow that enhances the AI's ability to tackle complex, high-stakes tasks with thoroughness and depth.
What it needs
About 4k tokens when loaded. Last updated 2026-08-07. 84,236 stars on the source repository.
What this skill does
/deep-think — Maximum Cognitive Effort Protocol Goal: Activate the highest level of cognitive architecture for complex, multi-step, or ambiguous tasks where quality matters more than speed. Time: 5-15 minutes depending on complexity tier. When to use: Architecture decisions, multi-system changes, irreversible actions, anything where being wrong is expensive. This is NOT /reason. /reason = "follow the reasoning engine checklist" (fast, lightweight). /deep-think = "maximum effort with structured thinking, research, gap analysis, multi-perspective debate, and adversarial review" (slow, thorough). --- Complexity Gate (decide first) Before starting, classify the task: Tier When Phases Used ------ ------ ------------- SIMPLE Quick but thorough — single-domain analysis, clear question Phases 1 + 5 only MEDIUM Research + analysis — multi-source synthesis, comparison Phases 1-5 COMPLEX Full adversarial — architecture decisions, multi-system changes, irreversible actions All 8 phases --- Phase 1: ORIENT (Memory + Context) Load ALL available context before forming any opinion. 1. Read project memory if available (e.g., MEMORY.md, SESSION.md, or status logs) 2. Read known pitfalls and gotchas for the target domain (e.g., knowledge/gotchas.md if present) 3. Read the most relevant local knowledge, READMEs, or chunk files 4. Query any active workspace memory system (e.g., Honcho or agentmemory) when recent session continuity is needed 5. Check conversation logs or history for prior attempts or context on this topic 6. Identify which projects, modules, or systems are involved If any tool fails: Note "degraded mode" and continue with available sources. Do NOT stall. --- Phase 2: QUESTION (Challenge the Request) Before solving, challenge the problem statement itself. 1. Use a sequential thinking or reasoning tool (such as sequential-thinking MCP, if available) to decompose the problem into component parts 2. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills nexu-io/deep-think-maximum-cognitive-effort-protocol-mq8kvw92