metacognitive-guard
Improve Complex Reasoning Escalation
Complex AI work can stall when uncertainty and tradeoffs are not recognized early. This skill gives Claude, Codex, and Claude Code a checklist for escalating difficult reasoning to deeper analysis.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "metacognitive-guard" from https://skillstore.io/skills/ancplua-metacognitive-guard.md and its manifest at https://skillstore.io/api/skills/ancplua-metacognitive-guard/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "metacognitive-guard". Should I use Redis or PostgreSQL for session storage?
Expected outcome:
- Escalation recommended because the decision has competing constraints and operational tradeoffs.
- The escalation should include current database use, traffic level, expiry needs, team skills, and cost limits.
Using "metacognitive-guard". Here is a draft answer with several caveats and no recommendation.
Expected outcome:
The skill flags hedging and lack of commitment. It recommends selecting one option with confidence, then asking targeted clarifying questions.
Using "metacognitive-guard". Do I need deeper analysis for a simple typo fix?
Expected outcome:
No escalation is suggested. The task has a narrow scope, clear success criteria, and low rework cost.
Security Audit
SafeAll static findings are false positives caused by Markdown inline code, YAML documentation blocks, and prose about decision ambiguity. I found no prompt injection, data exfiltration, credential access, filesystem access, network access, or executable command path in SKILL.md.
Risk Factors
⚙️ External commands (4)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/ancplua-metacognitive-guard/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/ancplua-metacognitive-guard?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/ancplua-metacognitive-guard?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/ancplua-metacognitive-guard/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/ancplua-metacognitive-guard.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
ANcpLua. (2026). metacognitive-guard security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/ancplua-metacognitive-guard/audits/9BibTeX citation
@techreport{ancplua-ancplua-metacognitive-guard-2026,
author = {ANcpLua},
title = {metacognitive-guard security audit report (audit version 9)},
institution = {Skillstore},
year = {2026},
number = {9},
url = {https://skillstore.io/skills/ancplua-metacognitive-guard/audits/9},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "metacognitive-guard security audit report (audit version 9)"
version: "unspecified"
type: report
authors:
- name: "ANcpLua"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/ancplua-metacognitive-guard/audits/9"
identifiers:
- type: other
value: "skillstore:ancplua-metacognitive-guard:audit:9"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Guide Architecture Decisions
Escalate architecture decisions when performance, cost, and maintainability constraints conflict.
Improve Assistant Answers
Check whether a response needs deeper analysis before giving a recommendation on ambiguous implementation choices.
Standardize Review Escalation
Define when teams should request deeper review for edge cases, integrations, or failure modes.
Try These Prompts
Use metacognitive-guard on this question. Tell me if the task needs deeper analysis before answering: [question]
Review my draft answer with metacognitive-guard. Identify struggle signals and rewrite it with a clear recommendation: [draft]
Apply metacognitive-guard to this architecture choice. Decide whether to escalate, then summarize context, constraints, success criteria, and the core question: [choice]
Use metacognitive-guard for a high-risk decision. Compare possible interpretations, list missing constraints, propose an escalation prompt, and define how to synthesize the result: [decision]
Best Practices
- Use the checklist before answering decisions with multiple valid approaches.
- Provide context, constraints, success criteria, and the exact decision when escalating.
- Synthesize partner analysis into a direct recommendation instead of forwarding it unchanged.
Avoid
- Escalating every task, including simple edits with clear success criteria.
- Using escalation to avoid making a recommendation when evidence is sufficient.
- Sending vague prompts that omit constraints, context, or the actual decision.