Skip to content
Get packs

Science Method Investigator

Diagnose or investigate a problem with a locked goal, competing testable hypotheses and decisive evidence.

Source Download skill
Repository
danielmiessler/LifeOS
Stars
19,682
Licence
MIT
Files
8 files

What it does

Once installed, the agent stops jumping to the first guess when you ask why something is happening. It first writes a clear goal, then proposes several competing explanations that could each be proven wrong, and picks the cheapest checks that tell them apart. Quick problems get a timeboxed diagnosis, while complex or research-grade questions get a full auditable record and a stopping decision. Requests to repeatedly tune a measured number are handed off to a separate HillClimb skill.

Great for

  • Developers chasing a flaky test or intermittent bug who want a disciplined debugging process
  • Analysts or researchers framing a complex question before collecting evidence
  • Anyone who wants explanations checked against evidence rather than accepted on instinct

Skip it if

  • Obvious five-minute fixes or creative/generative work, which the skill says need no ceremony (the process adds overhead)
  • Repeatedly optimising a measured metric; the skill routes this to HillClimb, so it is the wrong tool on its own
  • Multi-angle exploration of a topic; the skill points to IterativeDepth for that

Try asking

  • “Why is this test flaky?”
  • “Figure out why our API latency doubled last week.”
  • “Investigate the drop in signup conversion with a structured investigation.”
  • “Define the goal for our migration project before we start looking at evidence.”
Step-by-step guide: how to use Science Method InvestigatorInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 8 original files, the MIT licence and an attribution note.

    Science.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip Science.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip Science.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • State the problem and any evidence you already have up front so the agent can set a precise goal.
  • Ask for a QuickDiagnosis first; escalate to StructuredInvestigation only if the question stays unresolved.
  • Optional personal customizations can be placed in the documented USER/CUSTOMIZATIONS folder to override the defaults.

What's inside

  • Examples.md
  • Intents/DefineGoal.md
  • Intents/QuickDiagnosis.md
  • Intents/StructuredInvestigation.md
  • Methodology.md
  • Protocol.md
  • SKILL.md
  • Templates.md

Redistributed unmodified from danielmiessler/LifeOS under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.