Overthink

Chains 8 explicit reasoning modes (model, evaluate, decide, etc.) for ambiguous, high-stakes prompts

Tested · Works

Test report

Verdict
Tested · Works
Score
8.8/10
Tested
Jul 14, 2026
Environment
Claude Code 2.x (agent harness)
Upstream re-checked
Aug 10, 2026 · ab87487

Ran a real 'monolith vs microservices' decision baseline-vs-skill: both reached the same recommendation, but the skill run explicitly surfaced the unstated assumption (no confirmed scaling bottleneck) and stated a concrete condition that would flip the call -- baseline just gave a generic pros/cons list.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 5/5
  • Triggers reliably 5/5
  • Output vs. baseline 7/10
  • Docs & honesty 5/5

What Overthink does

Defines 8 thinking modes (model, create, diagnose, verify, evaluate, decide, taste, synthesize) and has the agent build a private todo ledger picking and chaining the modes relevant to the task, reading one reference file at a time. Triggers on complex reasoning, decision-making, nuance, evaluation or communication-sensitive work, and explicitly says not to activate itself for simple, non-manually-invoked tasks.

How to install Overthink

git clone https://github.com/s0md3v/overthink
cd overthink
mkdir -p ~/.claude/skills
cp -r skill ~/.claude/skills/overthink

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Overthink

  • /overthink Chains 8 explicit reasoning modes (model, evaluate, decide, etc.) for ambiguous, high-stakes prompts

It also activates on plain-language prompts like these:

  • Should we migrate our 15-person startup's monolith to microservices?
  • This decision is ambiguous and high-stakes, reason through it carefully
  • Evaluate and decide between these two vendor contracts for me

Frequently asked questions

Is the Overthink skill free?
Yes. The skill itself is free from s0md3v/overthink. SkillProof publishes the install command and an independent test verdict at no cost.
Does Overthink work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Jul 14, 2026. Verdict: Tested · Works. Ran a real 'monolith vs microservices' decision baseline-vs-skill: both reached the same recommendation, but the skill run explicitly surfaced the unstated assumption (no confirmed scaling bottleneck) and stated a concrete condition that would flip the call -- baseline just gave a generic pros/cons list.
What is the Overthink SkillProof Score?
8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 5/5.
How do I install Overthink?
Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Overthink with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.