Hum Socratic

What it does

When your agent uses it

• The user is explicitly facilitating learning
• Coaching a person through their own thinking
• Needs a structured questioning sequence to probe a held belief
• Even if they say play devils advocate on this claim

Example prompts

• “play devil”
• “how should I coach my team through this problem via questions”
• “/hum-socratic”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1Clarify the thinker's claim precisely
2. 2Probe assumptions — what must be true for the claim to hold?
3. 3Request evidence — what supports the belief? Is it sufficient?
4. 4Introduce counter-perspectives — what would a critic say?
5. 5Explore implications — if right, what follows? If wrong, what changes?
6. 6Synthesize (the IRON LAW checkpoint) — summarize, confirm understanding, then continue

What it can do on your machine

Read from SKILL.md and the folder at commit 4e7f4f8. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Hum Socratic loads about 1.3k tokens when it runs, and up to ~4.6k if it reads all its reference files. Until then it costs about 174 tokens; SKILL.md has 377 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~174

When it runs · the whole SKILL.md, loaded when a task matches

~1.3k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~4.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from asgard-ai-platform/skills at commit 4e7f4f8, republished under its MIT licence (© asgard-ai-platform). 377 words, ~1,259 tokens.

name

hum-socratic

description

Apply Socratic questioning — systematic inquiry via clarification, assumption-probing, evidence-testing, perspective-shifting, implication-tracing, and meta-questions — to coach learning or surface hidden assumptions in a person's reasoning. Use this skill when the user is explicitly facilitating learning, coaching a person through their own thinking, or needs a structured questioning sequence to probe a held belief, even if they say 'play devil's advocate on this claim' or 'how should I coach my team through this problem via questions'. Do NOT use for open-ended brainstorming, information gathering, or requirements-discovery question lists where no belief is being probed.

metadata.category

WP-19 文學院/人文

metadata.tags

humanities, socratic-method, questioning, pedagogy

IRON LAW: More Than 3 Consecutive Questions Without Summary Produces
Confusion, Not Insight

Agents applying Socratic questioning default to an unbounded chain of
questions. After ~3 questions, the thinker loses the thread — they can't
hold the question hierarchy in working memory. Pause every 2-3 questions
to SUMMARIZE what the thinker has revealed so far ("So your position is X
because Y, but you're unsure about Z — is that right?"). Then resume.
Without this checkpoint, the session feels like an interrogation, not a
guided inquiry.

Also: never ask a question whose answer you already embedded in the
phrasing. "Don't you think X is problematic?" is a leading assertion
disguised as a question. Rephrase as "What are the consequences of X?"

# Socratic Inquiry: {Topic}

## Starting Position
{The thinker's current belief or question}

## Question Sequence
1. [Clarification] {question}
   → Expected insight: {what this reveals}
2. [Assumption] {question}
   → Expected insight: {what this surfaces}
3. [Evidence] {question}
   → Expected insight: ...
4. [Perspective] {question}
   → Expected insight: ...
5. [Implication] {question}
   → Expected insight: ...

## Target Insight
{What the thinker should arrive at through this sequence}

© asgard-ai-platform, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in hum-socratic of asgard-ai-platform/skills.

• SKILL.md
• examples/sample_scenario.md
• references/facilitation-guide.md

Open the folder on GitHubat commit 4e7f4f8