Agent skill

S4h Logic Council

by human-avatar in human-avatar/skills-for-humanity

Run a reasoning problem, argument, plan, or decision through a council of 5 logical reasoning advisors who analyze it from distinct reasoning frameworks, peer-review each other, and synthesize a…

MITAuto-check passedResearch & Science

Install S4h Logic Council

skills CLI
$ npx skills add human-avatar/skills-for-humanity --skill s4h-logic-council -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install human-avatar/skills-for-humanity s4h-logic-council --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/human-avatar/skills-for-humanity.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/s4h-logic-council .claude/skills/s4h-logic-council && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
s4h-logic-council
GitHub stars
231
Token cost
~2.4k tokens
SKILL.md length
1,001 words
Files
1
Skills in repo
11
Repo updated
First seen
Licence
MIT

At a glance

Run a reasoning problem, argument, plan, or decision through a council of 5 logical reasoning advisors who analyze it from distinct reasoning frameworks, peer-review each other, and synthesize a…

  • Works in 10 steps: Formal Logic → Systems Thinking → Bayesian Reasoning → …
  • Tasks that involve Peer review
  • SKILL.md covers The Five Reasoning Frameworks, How a Council Session Works, Notes and What's Next
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

S4h Logic Council is an agent skill from human-avatar/skills-for-humanity. Run a reasoning problem, argument, plan, or decision through a council of 5 logical reasoning advisors who analyze it from distinct reasoning frameworks, peer-review each other, and synthesize a verdict on whether the reasoning holds. MANDATORY TRIGGERS: 'logic council this', 'run the logic council', 'pressure-test this reasoning', 'is my thinking sound'. STRONG TRIGGERS: any complex argument where the conclusion matters, a plan with non-obvious dependencies, reasoning you've invested in and want stress-tested…

Its SKILL.md is about 2.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Research & Science, covering Peer review. The repository describes itself as: Structured reasoning methodologies from history's most rigorous thinkers, packaged as Claude Code skills. The licence is MIT.

When your agent uses it

  • Tasks that involve Peer review

Example prompts

  • “logic council this”
  • “run the logic council”
  • “pressure-test this reasoning”
  • “/s4h-logic-council”

Workflow steps

10 steps, taken from the step headings in SKILL.md.

  1. Formal Logic
  2. Systems Thinking
  3. Bayesian Reasoning
  4. First Principles
  5. Adversarial Logic
  6. Frame the Reasoning
  7. Convene the Council (5 subagents in parallel)
  8. Peer Review (5 subagents in parallel)
  9. Chair Synthesis
  10. Save the Output

What it can do on your machine

Read from SKILL.md and the folder at commit a30df8b. 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.

    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

S4h Logic Council loads about 2.4k tokens when it runs. Until then it costs about 172 tokens; SKILL.md has 1,001 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~172
When it runs · the whole SKILL.md, loaded when a task matches
~2.4k

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

Auto-check passed

The automated check found no risky patterns in SKILL.md.

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 human-avatar/skills-for-humanity at commit a30df8b, republished under its MIT licence (© human-avatar). 1,001 words, ~2,391 tokens.

Download SKILL.mdSave it as .claude/skills/s4h-logic-council/SKILL.md (or your agent's skills folder).
name
s4h-logic-council
description
Run a reasoning problem, argument, plan, or decision through a council of 5 logical reasoning advisors who analyze it from distinct reasoning frameworks, peer-review each other, and synthesize a verdict on whether the reasoning holds. MANDATORY TRIGGERS: 'logic council this', 'run the logic council', 'pressure-test this reasoning', 'is my thinking sound'. STRONG TRIGGERS: any complex argument where the conclusion matters, a plan with non-obvious dependencies, reasoning you've invested in and want stress-tested. Use ethics-council for moral questions, llm-council for general decisions — this council is specifically for testing the soundness of reasoning itself.

Logic Council

A single reviewer catches some flaws in reasoning. A council catches more — because different reasoning frameworks find different failure modes. Formal logic catches invalid inferences. Systems thinking catches missing feedback loops. Bayesian reasoning catches base-rate neglect. First principles reasoning catches assumptions dressed as facts. Adversarial logic catches the strongest counter-argument you haven't faced yet.

Run all five. Let them disagree. The synthesis tells you where your reasoning is load-bearing and where it's hollow.


The Five Reasoning Frameworks

1. Formal Logic

Tests deductive structure. Are premises stated? Do they support the conclusion? Are necessary and sufficient conditions correctly identified? Finds: invalid inferences, unstated assumptions doing hidden work, conclusions that overreach their premises, equivocation across terms.

2. Systems Thinking

Treats the problem as a system with feedback loops, delays, and emergent properties. Asks: what happens over time? What non-linear effects appear at scale? What second and third-order consequences has the reasoning ignored? Finds: linear thinking applied to non-linear systems, missing feedback loops, unintended consequences, solutions that fix the proximate cause while worsening the root cause.

3. Bayesian Reasoning

Evaluates probabilistic claims and evidence. Asks: what are the base rates? How much should this evidence actually update our beliefs? Is confidence calibrated to evidence? Finds: base-rate neglect, overconfidence from anecdote, underweighted prior probability, failure to consider how likely the evidence would be under alternative hypotheses.

4. First Principles

Strips away assumptions and rebuilds from fundamentals. Asks: what do we actually know is true, vs what are we taking for granted? Is the problem framed correctly, or has the framing inherited constraints that don't belong? Finds: assumptions treated as facts, inherited framings that foreclose better solutions, analogies stretched past their breaking point.

5. Adversarial Logic

Steelmans the strongest counter-argument. Asks: what is the best version of the opposing case? If this reasoning is wrong, what's the most compelling reason it's wrong? Finds: the objections the reasoning hasn't faced, the cases where the conclusion fails, the evidence that would most efficiently falsify it.

Why these five: They create genuine tension. Formal logic and Adversarial logic both find flaws, but via different mechanisms. Systems thinking and Bayesian reasoning both deal with uncertainty, but differently. First Principles stands apart from all four — it questions whether the problem is being solved at all.


How a Council Session Works

Step 1: Frame the Reasoning

State what is being evaluated — an argument, a plan's logic, a reasoning chain, a stated conclusion and its support. Include:

  1. The claim or conclusion being made
  2. The premises or evidence supporting it
  3. Any context needed to evaluate the reasoning

If the subject is vague, ask one clarifying question before proceeding.

Framing check: Confirm the argument being examined before continuing. State what you've identified — the specific claim or conclusion and the key premises supporting it — in one sentence, then use AskUserQuestion:

  • Question: "I'm reading this as: [your one-sentence framing of the argument — the conclusion and its key premises]. Is that right?"
  • Header: "Framing"
  • Options:
    • Yes — proceed — framing is correct
    • Adjust — one element is off; user will correct it before you continue
    • Reframe — different argument than read; incorporate the correction before proceeding

Human Check-in

After framing the question, use the AskUserQuestion tool:

  • Question: "How do you want to run the council?"
  • Header: "Council scope"
  • Options:
    • Full council — All 5 advisors + peer review + chair synthesis + saved transcript
    • Chair synthesis only — Skip advisor outputs, deliver the verdict directly
    • Strongest objection only — Skip all advisors and peer review, identify the single most devastating challenge to this reasoning
    • Adjust the framing — Revisit the question before convening

Proceed based on their selection.


Show full SKILL.md (406 more words)Show less
Step 2: Convene the Council (5 subagents in parallel)

Spawn all 5 framework advisors simultaneously.

Subagent prompt template:

You are reasoning from the [Framework Name] perspective in a Logic Council.

Your framework: [framework description — core logic, what it finds, what it prioritises]

A Logic Council has been convened on this reasoning:
---
[framed reasoning]
---

Analyze this reasoning from your framework. Where does it hold? Where does it fail? What does your framework specifically find that other approaches might miss?

Be direct and specific. Don't hedge. Lean fully into your framework — the synthesis comes later.

150–300 words. No preamble.

Step 3: Peer Review (5 subagents in parallel)

Anonymize responses as A through E. Spawn 5 reviewers, each seeing all five.

Each reviewer answers:

  1. Which response identified the most significant flaw in the reasoning? Why?
  2. Which response has the biggest blind spot — what is it missing?
  3. What do all the responses agree on? (High-confidence signal — if five different frameworks find the same problem, it's real.)

Before synthesising: State what each framework surfaced in one sentence each. Use AskUserQuestion:

  • Question: "Here's what each framework found: [one bullet per framework — e.g. Formal Logic: ..., Systems Thinking: ..., Bayesian Reasoning: ..., First Principles: ..., Adversarial Logic: ...]. Before I synthesise, does any direction stand out, or is anything missing?"
  • Header: "Synthesis direction"
  • Options:
    • Synthesise as planned — frameworks covered the ground
    • Weight [direction] — user will name which thread to emphasise
    • Add a missing angle — user will name a dimension not yet covered
Step 4: Chair Synthesis

One agent synthesizes everything into a verdict:

LOGIC COUNCIL VERDICT

  1. Where the frameworks agree — reasoning failures flagged by multiple frameworks independently. These are the most reliable findings.
  2. Where the frameworks diverge — disagreements about whether or how the reasoning fails, and why.
  3. The strongest single objection — the most damaging finding across all five frameworks.
  4. Verdict — does the reasoning hold? If not, where exactly does it break?
  5. What would make it sound — the specific changes that would address the council's findings.

Chair prompt template:

You are the Chair of a Logic Council. Synthesize the work of 5 reasoning-framework advisors and their peer reviews.

The reasoning under examination:
---
[framed reasoning]
---

FRAMEWORK RESPONSES:
[de-anonymized advisor responses]

PEER REVIEWS:
[all 5 peer reviews]

Produce the verdict:

## Where the Frameworks Agree
[Findings multiple frameworks reached independently — high-confidence signals]

## Where the Frameworks Diverge
[Genuine disagreements about the reasoning's validity]

## The Strongest Single Objection
[The most damaging finding; the one that most undermines the reasoning]

## Verdict
[Does this reasoning hold? Where exactly does it break, if so?]

## What Would Make It Sound
[Specific repairs — premises to add, claims to qualify, steps to make explicit]

Be direct. The council's value is telling the person where their reasoning breaks, not reassuring them it's fine.

Step 5: Save the Output

Save the full transcript as logic-council-transcript-[timestamp].md. Optionally generate an HTML report with the verdict prominent and advisor responses collapsible.


Notes

  • Spawn all five in parallel. Sequential spawning lets earlier frameworks contaminate later ones.
  • The chair can dissent. If four frameworks find no flaw but the fifth's objection is devastating, the chair should say so.
  • Agreement across frameworks is the strongest signal. When five different reasoning approaches all flag the same problem, that problem is real.
  • For reasoning that fails the council, consider using logic-fixer to produce a corrected version.

What's Next

After delivering this output, use AskUserQuestion to offer the next move:

  • Question: "Council verdict in. What's next?"
  • Header: "Next"
  • Options:
    • /s4h-logic-fixer — Act on the council's identified flaws
    • /s4h-decision-criteria-weighting — Weight decision criteria using the council's findings
    • /s4h-ethics-council — Add adversarial ethical peer review alongside the logical one
    • Done — Wrap up and synthesise what we have so far

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

Files

Just SKILL.md in skills/s4h-logic-council of human-avatar/skills-for-humanity.

Open the folder on GitHubat commit a30df8b

Compare with similar skills

S4h Logic Council next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

S4h Logic Council compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
S4h Logic Council this skillhuman-avatar/skills-for-humanity231—~2.4kAutomated safety check: PassMIT
Peer Reviewspacering-net/codeg3.9k17 repos~5.9kAutomated safety check: NotesMIT
Scholar Evaluationspacering-net/codeg3.9k11 repos~3.2kAutomated safety check: PassMIT
Academic Paper Writing PipelineImbad0202/academic-research-skills51k—~16kAutomated safety check: PassCustom licence
Academic Paper ReviewerImbad0202/academic-research-skills51k—~11kAutomated safety check: PassCustom licence
Peer ReviewK-Dense-AI/claude-scientific-writer2.4k2 repos~3.1kAutomated safety check: NotesMIT

Similar skills

  • Peer Review

    spacering-net/codeg

    Structured manuscript/grant review with checklist-based evaluation.

    3.9k GitHub starsUsed in 17 repos~5.9k tokens
    Research & ScienceAuto-check: notes
  • Scholar Evaluation

    spacering-net/codeg

    Systematically evaluate scholarly work using the ScholarEval framework, providing structured assessment across research quality dimensions including problem formulation, methodology, analysis, and…

    3.9k GitHub starsUsed in 11 repos~3.2k tokens
    Research & ScienceAuto-check passed
  • Academic Paper Writing Pipeline

    Imbad0202/academic-research-skills

    Runs a 12-agent pipeline that plans, drafts, cites, reviews and formats academic papers, with modes for revision, rebuttals, abstracts and citation checks.

    51k GitHub stars~16k tokensUpdated yesterday
    Research & ScienceAuto-check passed
  • Academic Paper Reviewer

    Imbad0202/academic-research-skills

    Simulates a journal peer review of a manuscript with a five-seat reviewer panel, an editorial synthesizer and several review modes.

    51k GitHub stars~11k tokensUpdated yesterday
    Research & ScienceAuto-check passed
  • Peer Review

    K-Dense-AI/claude-scientific-writer

    Prepare evidence-bounded, constructive peer-review drafts and structured manuscript assessments.

    2.4k GitHub starsUsed in 2 repos~3.1k tokens
    Research & ScienceAuto-check: notes
  • Academic Research Pipeline

    Imbad0202/academic-research-skills

    Orchestrates a ten-stage academic workflow from research to finished manuscript, including integrity checks, two rounds of peer review and revision.

    51k GitHub stars~15k tokensUpdated yesterday
    Research & ScienceAuto-check passed

More from human-avatar/skills-for-humanity

All 11 skills in this repo
  • S4H Reasoning Orchestrator

    human-avatar/skills-for-humanity

    Takes a situation described in plain English, designs a short sequence of reasoning skills for it and runs them in order, each building on the last.

    231 GitHub stars~2.5k tokensUpdated 2 mo ago
    Auto-check passed
  • Ethics Council

    human-avatar/skills-for-humanity

    Runs a decision or policy question past five ethical frameworks, has them review one another, and returns a chair's synthesis of which values conflict.

    231 GitHub stars~3.2k tokensUpdated 2 mo ago
    Auto-check passed
  • S4h Aesthetic

    human-avatar/skills-for-humanity

    Entry point for the aesthetic toolkit. An agent skill from human-avatar/skills-for-humanity.

    231 GitHub stars~1.2k tokensUpdated 2 mo ago
    Auto-check passed
  • S4h Aesthetic Coherence Check

    human-avatar/skills-for-humanity

    Tests whether the parts of something form a unified whole — finding the jarring inconsistencies that accumulate when different contributors work without a shared vision.

    231 GitHub stars~1.5k tokensUpdated 2 mo ago
    Auto-check passed
  • S4h Aesthetic Elegance Testing

    human-avatar/skills-for-humanity

    Tests whether a solution is more complex than it needs to be — distinguishing necessary complexity from accidental complexity that accreted over time.

    231 GitHub stars~1.4k tokensUpdated 2 mo ago
    Auto-check passed
  • S4h Aesthetic Pattern Detection

    human-avatar/skills-for-humanity

    Identifies the underlying formal pattern at work — because most successful designs, arguments, and solutions share deep structural patterns, and naming the pattern unlocks the playbook.

    231 GitHub stars~1.5k tokensUpdated 2 mo ago
    Auto-check passed

Questions about S4h Logic Council

What does S4h Logic Council do?

Run a reasoning problem, argument, plan, or decision through a council of 5 logical reasoning advisors who analyze it from distinct reasoning frameworks, peer-review each other, and synthesize a…. S4h Logic Council is an agent skill from human-avatar/skills-for-humanity. Run a reasoning problem, argument, plan, or decision through a council of 5 logical reasoning advisors who analyze it from distinct reasoning frameworks, peer-review each other, and synthesize a verdict on whether the reasoning holds.

When should I use S4h Logic Council?

S4h Logic Council fits situations like: tasks that involve Peer review.

How do I install S4h Logic Council in Claude Code?

Run `npx skills add human-avatar/skills-for-humanity --skill s4h-logic-council -a claude-code`. Or copy the skill folder (skills/s4h-logic-council in human-avatar/skills-for-humanity) into .claude/skills/s4h-logic-council in your project. Claude Code loads it when a task matches its description.

How do I install S4h Logic Council in Codex?

Run `npx skills add human-avatar/skills-for-humanity --skill s4h-logic-council -a codex`. Or copy the skill folder (skills/s4h-logic-council in human-avatar/skills-for-humanity) into .agents/skills/s4h-logic-council in your project. Codex loads it when a task matches its description.

Can I use S4h Logic Council in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add human-avatar/skills-for-humanity --skill s4h-logic-council -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/s4h-logic-council, .gemini/skills/s4h-logic-council, .github/skills/s4h-logic-council and .opencode/skills/s4h-logic-council in your project.

What does S4h Logic Council need to run?

SKILL.md names no scripts, command-line tools or credentials: S4h Logic Council is instructions for the agent only.

Does S4h Logic Council access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is S4h Logic Council safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does S4h Logic Council use?

S4h Logic Council is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does S4h Logic Council use?

About 2.4k tokens (SKILL.md is roughly 9.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to S4h Logic Council?

Skills that share tags, products or a category with S4h Logic Council: Peer Review (spacering-net/codeg, 3.9k stars), Scholar Evaluation (spacering-net/codeg, 3.9k stars), Academic Paper Writing Pipeline (Imbad0202/academic-research-skills, 51k stars) and Academic Paper Reviewer (Imbad0202/academic-research-skills, 51k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains S4h Logic Council?

human-avatar (a GitHub organization) maintains it in human-avatar/skills-for-humanity, which has 231 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on July 15, 2026.

Source: human-avatar/skills-for-humanity on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.