How to use the Rational Buddhism Verifier agent skill

What does the rational-buddhism skill make the agent do?

The skill does not add a tool or connect to anything. It changes how the agent reasons about a claim. Its core idea is to try everything, test it yourself, stay skeptical, keep what is useful and discard what is not. Claims that cannot be falsified do not become basic truths for you.

In practice the agent asks whether a claim makes predictions that could be proven wrong, and whether you would notice if it failed. Claims that fail this test are marked as not basic truths. Inner techniques such as meditation, observation and acceptance can be kept, and the agent suggests a way to test them on yourself.

• Asks whether the claim makes falsifiable predictions
• Suggests a personal trial for practices you can experience, such as meditation or fasting
• Sorts results into keep (repeatable benefit) and discard (no effect or cannot be checked)
• Helps you write a short personal list of beliefs with the evidence level for each

How do you install the skill?

A skill is a folder of instructions that an agent loads when it is relevant. The rational-buddhism skill is distributed as a ZIP file from this directory page. Download the ZIP and unzip it. The folder contains SKILL.md, test-prompts.json and test-results.md.

Choose where to put it. For a personal install, place the unzipped folder in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder. Alternatively, in the Claude apps, open Settings, go to Capabilities and upload the skill under Skills.

• Personal install: unzip into ~/.claude/skills/
• Project install: unzip into the project's .claude/skills/
• Claude apps: Settings → Capabilities → Skills, then upload

How do you trigger the skill with natural requests?

You do not need to name the skill. The agent is meant to use it when your question is about whether a claim should be believed, how to set up your own standard for belief, or how science and spiritual practice relate. The skill's trigger words include 该不该信, 玄学, 灵修, 冥想, 可证伪, 验证, 理性 and the English words rational buddhism, verify and belief.

Plain requests work well. Describe the claim and ask what the evidence is, or ask for a way to test it yourself.

• 「这个该信吗」 (is this worth believing?)
• 「怎么验证一个说法」 (how do I verify a claim?)
• 「科学和冥想冲突吗」 (do science and meditation conflict?)
• 「怎么亲测又不走火入魔」 (how do I test it myself without going overboard?)
• "is this true / falsifiable / rational approach to spirituality"

What does a typical exchange look like?

Suppose you ask whether an energy-healing session is worth believing. The skill directs the agent to first ask whether the claim makes predictions that could fail. If it cannot, the agent treats the claim as not a basic truth. If the session itself is something you can experience, the agent can suggest a trial in which you record how you feel before and after, and then keep the parts that work for you.

The source example is a reader who first dismissed an energy practice as nonsense, then tried it with an open mind while meditating lying down, and found it pleasant. The skill's point is that testing comes before prejudice in either direction, while the claim itself is still judged by whether it can be checked.

What are the limits and safety notes?

The skill requires nothing to be installed beyond the files. It does not run scripts and does not use the network. It is a reasoning framework, so its output is only as good as the evidence you give it and your own careful testing.

Do not use it in these cases. It is not for someone grieving or dying who wants religious or spiritual comfort; the skill says it should not push a rationalist view on a personal faith choice. It is not for judging literature, love or meaning, which cannot be assessed by falsifiability. It also does not check clinical evidence, so medical decisions should stay with a doctor.

• Not for end-of-life or grief situations that call for faith-based comfort
• Not for literature, love or meaning questions
• Does not check clinical evidence; do not use it in place of medical advice
• The skill's own author notes that the verification standard is itself partly subjective, and that the rationalist framework explains non-rational needs only in limited ways

Tips for getting useful results

Give the agent the exact claim, the source where you heard it and what you would expect to see if it were true. Specific claims can be tested; vague ones usually cannot. Ask the agent to name what would prove the claim wrong.

Keep a short list of the beliefs you adopt, with a note on the evidence behind each. The skill suggests three such entries as a starting point. Pair it with the judgment-training and monkey-mind-meditation skills if you have them, since the related-skills notes describe them as dependencies and companions.

• Name the specific claim and what would count as failure
• Record before-and-after notes for any practice you try
• Keep a written list of beliefs with their evidence levels