awesome-skills/first-principles-skill

Selection

Select when

• You want Claude Code to challenge inherited architecture or design assumptions.
• A proposed solution appears more complex than the underlying problem.
• A decision is justified mainly by industry convention or another company's example.
• You need a structured record of assumptions, ground truths, and trade-offs.
• You are making a consequential foundational decision with long-term effects.
• You want English or Chinese trigger phrases for the same reasoning workflow.
• You need to derive a minimal solution before considering optional complexity.

Boundaries

Avoid when

• You need executable analysis, code transformation, or automated testing.
• You need verified facts without supplying evidence or domain expertise.
• You need a guaranteed optimal decision rather than a reasoning framework.
• The decision is routine and does not justify a multi-phase analysis.
• You require a documented standard Claude Code installation path without SuperClaude-related assumptions.
• You need orchestration across multiple agents, tools, or skills.
• You need a maintained library covering many reasoning methods.

Strengths

Capabilities

Risk profile

Risks and limitations

• Uncertainty: only README.md and SKILL.md were supplied, so the referenced case studies, example, LICENSE text, and complete repository tree were not inspected.
• Uncertainty: the README documents installation under a SuperClaude plugin path, but no SuperClaude prerequisite, manifest, or standard Claude Code skill-directory alternative is explained.
• Uncertainty: automatic triggering is documented, but no trigger evaluation or Claude Code compatibility test was supplied.
• Inference: early-stage maturity is suggested by version 0.2.0, the small repository size, and the absence of supplied tests or release evidence.
• The method depends on model judgment to distinguish facts from assumptions and can still produce unsupported ground truths.
• The skill provides procedural guidance rather than evidence gathering, calculation, experimentation, or formal verification.
• Referenced complementary skills are not bundled or linked to implementations in the supplied files.
• No minimum Claude Code version, update procedure, compatibility matrix, security policy, or telemetry statement was supplied.
• Repository metadata dates extend beyond the analysis context, so recency and adoption signals cannot be independently reconciled.
• The SpaceX and Tesla reference content was not supplied, so its sourcing, accuracy, and relevance are uncertain.

The supplied skill is declarative Markdown and evidences no scripts, tools, hooks, network calls, or credential handling. Its primary security risk is analytical: incorrect model-generated assumptions or recommendations could influence consequential technical decisions. No security policy, threat model, audit, or integrity checks were supplied.