Skip to content
Agentmart
Submit a tool

eureka-skill

Coordinate rigorous invention, mathematical research, formalization, validation, and high-integrity computation through a compact evidence-first skill…

Skill by minagayid

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/minagayid-eureka-skill

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Coordinate rigorous invention, mathematical research, formalization, validation, and high-integrity computation through a compact evidence-first skill collection. Use for complex breakthrough, theory, invention, optimization, or computational-research requests that span several specialist workflows and need explicit routing, reproducibility, and proof-gap controls.

Use cases: research

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/eureka-skill

What's inside

Skill name
eureka-skill
Description
Coordinate rigorous invention, mathematical research, formalization, validation, and high-integrity computation through a compact evidence-first skill collection. Use for complex breakthrough, theory, invention, optimization, or computational-research requests that span several specialist workflows and need explicit routing, reproducibility, and proof-gap controls.
License
MIT
Instructions
34 lines in SKILL.md · view on GitHub

Files (4)

  • SKILL.md
  • catalog.md · references/
  • routing.md · references/
  • eureka-program.md · templates/

SKILL.md (first lines)

# eureka_skill

Use **eureka_skill** as a routing layer for ambitious interdisciplinary work. It does not replace specialist skills or convert an attractive idea into a theorem, patent, safe design, or commercial product. It selects the smallest route that can establish the next bounded fact.

## Eureka workflow

1. **Frame the decision.** Define target, system boundary, deliverable, risk/cost of error, success metric, prohibited claims and stop condition.
2. **Build an evidence ledger.** Label every material statement sourced, computed, inferred, proposed or unresolved. Preserve source/artefact links and uncertainty.
3. **Select the minimum route.** Read [routing.md](references/routing.md) and choose only the specialist skills needed for the current gate.
4. **Create mechanisms, not slogans.** Require baseline, conserved/budgeted quantities, causal chain, counterfactual/negative control, predicted signature and falsifier.
5. **Formalize and compute responsibly.** Freeze inputs, units, data/seed policy, numerical/formal method, verification, validation, uncertainty, manifest and rerun command before execution.
6. **Run an adversarial gate.** Check alternatives, boundary/degenerate cases, sensitivity, hidden conditions, decision reversal and cost of error. Escalate high-consequence sensitive results to independent implementation or measurement.
7. **Deliver a bounded handoff.** State what was established, by which versioned artefact, over which regime, what remains a proof gap, and the next reversible action.

## Core collection

Read [catalog.md](references/catalog.md) when selecting a specialist or auditing the collection’s upgrade coverage.

| Layer | Skills |
|---|---|
| Framing and evidence | `critical-creative-thinking`, `research-synthesis-hypothesis`, `frontier-research-architect` |
| Discovery and mechanism | `innovation-breakthrough-design`, `contradiction-mechanism-lab`, `cross-domain-recombination`, `invention-generation-orchestrator` |
| Formalization and validation | `conjecture-formalization-lab`, `invention-validation-lab`, `bounded-invention-engineering` |
| Mathematical and super computation | `advanced-mathematics-computation`, `math-research-orchestrator`, `eureka-computational-lab`, `eureka-design-space-lab`, `eureka-formal-logic-lab` |

## Default deliverable

Start from [eureka-program.md](templates/eureka-program.md). Do not claim a result exceeds the strongest completed evidence layer. For instance, a finite numerical pass is finite computed evidence; an exhaustive search is only as strong as its coverage proof; a converged simulation is not automatically a validated model.

## Completion gate

Finish only with a decision, evidence classification, versioned artefact locations, reproducibility/VVUQ status where applicable, proof-gap ledger and next bounded action. Stop for user approval before irreversible external action.

From minagayid/innovator, MIT.

More from minagayid/innovator

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 3, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from minagayid

Cite this page

eureka-skill. Agent Mart. https://agentmart.dev/skills/minagayid-eureka-skill (last updated 2026-10-24)