How do I install the Invariant Finder skill?

Run npx skillmds@latest add 45ck/invariant-finder in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Invariant Finder skill do?

invariant-finder It is listed under Coding & Dev Tools on SkillMD.

Is Invariant Finder safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Invariant Finder?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Invariant Finder free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Invariant Finder?

45ck (@45ck) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Invariant Finder by 45ck — Agent Skill

Name: Invariant Finder
Author: 45ck

File contents

invariant-finder

Purpose

Identify invariants, preservation conditions, and breakpoints that should hold across operations or iterations.

Trigger this skill when

You need a structure or algorithm recommendation grounded in the real workload.
A current implementation choice feels arbitrary, costly, or hard to justify.
You want to reason about correctness or performance before writing more code.

Expected inputs

algorithm or data structure operations
state transitions
correctness goals
loop/recursion points
failure examples if any

Deliverables

candidate invariants
where they are established
where they are preserved or broken
correctness notes
test/proof prompts

Operating procedure

Identify the data or algorithmic concern this skill is meant to assess.
Separate facts about operations, sizes, ordering, and mutation from assumptions.
Review the current or proposed approach against the workload, invariants, complexity, and representation constraints.
Surface concrete risks, tradeoffs, and improvement options.
Prefer the least complex structure or algorithm that meets the stated needs.
Hand off to the next most relevant skill if the issue is broader than this skill alone can resolve.

Quality gates

Findings are tied to the stated workload or concrete algorithm steps.
Recommendations explain why the choice fits the context.
Tradeoffs are stated when multiple options are viable.
Output distinguishes asymptotic concerns from practical implementation risks.

Handoff targets

algorithm-trace-checker
complexity-red-flag-detector
data-structure-selector

Output style

Be explicit about uncertainty.
Prefer short, evidence-based findings over generic claims of efficiency.
Call out hidden assumptions such as sorted input, bounded size, or low mutation.
End with recommended next actions.

Failure modes to avoid

Do not recommend advanced structures without a workload that justifies them.
Do not rely on Big-O alone when locality or constant factors dominate.
Do not conflate proof of correctness with a single successful example.
Do not ignore edge cases, degenerate inputs, or mutation semantics.

Minimum output skeleton

## Summary
## Findings
## Risks
## Recommendations
## Tradeoffs
## Recommended next skill

---name: invariant-finderdescription: invariant-finder---# invariant-finder## PurposeIdentify invariants, preservation conditions, and breakpoints that should hold across operations or iterations.## Trigger this skill when- You need a structure or algorithm recommendation grounded in the real workload.- A current implementation choice feels arbitrary, costly, or hard to justify.- You want to reason about correctness or performance before writing more code.## Expected inputs- algorithm or data structure operations- state transitions- correctness goals- loop/recursion points- failure examples if any## Deliverables- candidate invariants- where they are established- where they are preserved or broken- correctness notes- test/proof prompts## Operating procedure1. Identify the data or algorithmic concern this skill is meant to assess.2. Separate facts about operations, sizes, ordering, and mutation from assumptions.3. Review the current or proposed approach against the workload, invariants, complexity, and representation constraints.4. Surface concrete risks, tradeoffs, and improvement options.5. Prefer the least complex structure or algorithm that meets the stated needs.6. Hand off to the next most relevant skill if the issue is broader than this skill alone can resolve.## Quality gates- Findings are tied to the stated workload or concrete algorithm steps.- Recommendations explain why the choice fits the context.- Tradeoffs are stated when multiple options are viable.- Output distinguishes asymptotic concerns from practical implementation risks.## Handoff targets- algorithm-trace-checker- complexity-red-flag-detector- data-structure-selector## Output style- Be explicit about uncertainty.- Prefer short, evidence-based findings over generic claims of efficiency.- Call out hidden assumptions such as sorted input, bounded size, or low mutation.- End with recommended next actions.## Failure modes to avoid- Do not recommend advanced structures without a workload that justifies them.- Do not rely on Big-O alone when locality or constant factors dominate.- Do not conflate proof of correctness with a single successful example.- Do not ignore edge cases, degenerate inputs, or mutation semantics.## Minimum output skeleton```md## Summary## Findings## Risks## Recommendations## Tradeoffs## Recommended next skill```

Frequently asked questions

Invariant Finder