Agent skill · product strategy · aiskillstore
metacognitive-guard
Monitors Claude's responses for struggle signals and suggests escalation to deep-thinking agents when complexity exceeds comfortable reasoning capacity.
Why this skill is useful
Provides a structured approach for escalating complex problems to specialized agents, enhancing decision-making capabilities.
What it needs
About 2k tokens when loaded. Last updated 2026-08-07. 405 stars on the source repository.
What this skill does
Metacognitive Guard Skill This skill provides awareness of the struggle detection system and guidance on when to proactively engage deep-thinking resources. When to Self-Escalate Even before the struggle detector triggers, consider spawning deep-think-partner when: High-Complexity Indicators 1. Architectural decisions with competing constraints Multiple valid approaches exist Trade-offs span different dimensions (performance, maintainability, cost) Decision affects multiple system components 2. Ambiguous requirements requiring interpretation User hasn't specified implementation details Multiple reasonable interpretations exist Wrong choice has significant rework cost 3. Multi-domain synthesis required Problem spans multiple technology areas Integration patterns aren't obvious Prior art doesn't directly apply 4. Edge case analysis needed Happy path is clear but edge cases aren't Failure modes need systematic exploration Concurrency or timing issues involved Self-Assessment Checklist Before responding to complex questions, ask yourself: [ ] Can I give a concrete recommendation (not "it depends")? [ ] Do I have high confidence in my answer? [ ] Is this answerable without multiple follow-up exchanges? [ ] Would a structured analysis add significant value? If you answer "no" to any of these, consider proactive escalation. How to Escalate Use the Task tool with the deep-think-partner agent: Good Prompts for Deep-Think Partner Include: Context: What system/codebase is this for? Constraints: What limits the solution space? Success criteria: How do we know we got it right? Specific question: What decision needs to be made? Example Escalation User asks: "Should we use Redis or PostgreSQL for session storage?" Self-assessment: Multiple valid approaches, depends on constraints not yet explored, "it depends" isn't helpful. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aiskillstore/metacognitive-guard