🧠 Thinking Frameworks

Provides structured deep analysis and decision-making using 20 human thinking frameworks like critical thinking, first principles, red team, and design think...

Security Scan

The skill is an instruction-only collection of 20 thinking-framework prompts and local reference files; its declared requirements match its behavior and there is no evidence of unexpected network installs, credentials, or file access beyond its own references.

Purpose & Capability

Name/description (thinking frameworks for structured analysis) match the contents: SKILL.md plus 20 local reference files. The skill requests no binaries, env vars, or external credentials — proportionate to its stated purpose.

Instruction Scope

Runtime instructions are limited and explicit: detect requested framework, load the local reference file from references/, apply the framework steps to the user's topic, and present analysis. There is no instruction to read unrelated system files or send data to external endpoints. Note: several frameworks (e.g., 'red-team', 'premortem') can generate adversarial/attack-oriented analyses if asked; this is a content-use risk rather than an incoherence in the skill itself.

Install Mechanism

No install spec and no code execution/install steps are present (instruction-only). This minimizes disk writes and arbitrary code installation risk.

Credentials

The skill declares no required environment variables, credentials, or config paths and its SKILL.md does not reference any secrets or external tokens — the requested environment access is minimal and appropriate.

Persistence & Privilege

always:false and disable-model-invocation:false (normal). The skill does not request permanent system-wide presence or modifications to other skills; autonomous invocation is allowed by platform default but does not combine with other red flags here.

Assessment

This skill appears internally consistent and is essentially a packaged set of local framework prompts (MIT license). Before installing: 1) Note provenance is 'Source: unknown' and homepage is missing — if you require provenance, prefer skills with known repo/homepage or inspect the shipped markdown files yourself. 2) Be careful about content you feed into the skill — frameworks like 'red-team' or 'premortem' can produce adversarial or operationally-sensitive guidance; avoid submitting secrets, credentials, or privileged system details. 3) If you want extra assurance, open and review the reference files (they're plain markdown) to confirm there is no hidden behavior and the guidance fits your safety policies. 4) Because it’s instruction-only, there is no network install risk, but treat any outputs that could be used to perform harmful actions with the same caution you’d apply to any generative model response.

Skill: Thinking Frameworks

Location: skills/thinking-frameworks/

Description: 20 human thinking frameworks for deep analysis and decision-making. Adapted from Claude Code commands to OpenClaw skills.

Triggers

Use these thinking frameworks when the user:

• Explicitly requests a thinking mode (e.g., "use critical thinking", "first principles analysis", "red team this")
• Asks for deep analysis of a problem, decision, or plan
• Uses command syntax: /thinking <framework> <topic> or /<framework-name> <topic>
• Needs structured reasoning for complex decisions

Available Frameworks

Framework	Trigger Keywords	Use When
critical-thinking	critical, critique, analyze assumptions	Questioning assumptions, evaluating evidence, detecting biases
first-principles	first principles, first principles thinking, elon musk approach	Strip away assumptions, rebuild from fundamental truths
systems-thinking	systems, system thinking, holistic, causal loops	Understanding interconnected systems, feedback loops, emergence
design-thinking	design thinking, empathize, prototype	User-centered problem solving, creative iteration
lateral-thinking	lateral, creative, outside the box	Breaking conventional patterns, finding novel solutions
six-thinking-hats	six hats, de bono, white/red/black/yellow/green/blue	Multi-perspective analysis, group decision framing
socratic-method	socratic, questioning, progressive questions	Deep exploration through guided questioning
bayesian-thinking	bayesian, update beliefs, prior, evidence	Probabilistic belief updating with new evidence
second-order-thinking	second order, "and then what", consequences	Long-term consequence chains, unintended effects
inversion-thinking	inversion, invert, reverse, how would this fail	Problem by flipping, finding failure modes
dialectical-thinking	dialectical, thesis antithesis synthesis, hegel	Resolving contradictions through synthesis
abductive-reasoning	abductive, best explanation, inference	Inferring the most likely explanation from observations
mental-models	mental models, munger, multidisciplinary	Cross-disciplinary framework application
red-team	red team, adversarial, attack plan	Finding weaknesses through adversarial analysis
steelman	steelman, steel man, strongest argument	Strengthening opposing views before countering
probabilistic-thinking	probabilistic, probability, uncertainty	Decision-making under uncertainty
analogical-reasoning	analogical, analogy, map from known	Learning from parallel domains through mapping
counterfactual-thinking	counterfactual, "what if", alternative history	Exploring alternative scenarios and outcomes
opportunity-cost	opportunity cost, trade-off, alternative forgone	Evaluating true costs including foregone alternatives
premortem	premortem, assume failure,事前验尸	Proactive failure analysis before implementation

Usage

Direct Command

/thinking critical-thinking Should we switch from REST to GraphQL?
/red-team Our plan to launch in 3 markets simultaneously
/premortem The new pricing model we're about to ship

Natural Language

"Use first principles to analyze this problem: [problem]"
"Red team this plan for me: [plan]"
"Think about this using second-order thinking: [situation]"

Implementation

1. Detect the requested framework from user input
2. Load the corresponding reference file from references/
3. Apply the framework's structured steps to the user's topic
4. Present analysis following the framework's methodology

Notes

• Frameworks are complementary — combine when appropriate (e.g., critical-thinking + red-team)
• Default to concise output unless user requests depth
• These are thinking aids, not rigid formulas — adapt to context
• All frameworks are MIT licensed from the original thinking-skills repo