A structured debate command that has an attacker challenge an idea, task, or decision and an adjudicator assess the discussion.

Use it to stress-test plans, work items, and decisions, provided you supply an existing work ID when required.

It helps expose weaknesses and risks that may be missed when reviewing an idea from only one perspective.

thoth:argue — Command by SeeleAI

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/seeleai/thoth/argue

Clone the repo

git clone --depth 1 https://github.com/SeeleAI/Thoth

Per session 25 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 788 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00025	$0.00788
Opus 5	$0.00013	$0.00394
Sonnet 5	$0.00005	$0.00158
Haiku 4.5	$0.00003	$0.00079

Measured 2d ago against content hash d7c1f384fd74, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

thoth:argue scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/argue.md · 67 lines

How it starts

The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/thoth:argue

Generated Surface

This file is generated from thoth.command_specs.COMMAND_SPECS. Do not hand edit.

Real Runtime Execution

The repo-local Thoth runtime command for this slash command has already been executed before Claude sees this prompt.

THOTH_ARGUE_ARGUMENTS_FILE="$(mktemp -t thoth-argue-arguments.XXXXXX)"
trap 'rm -f "$THOTH_ARGUE_ARGUMENTS_FILE"' EXIT
cat > "$THOTH_ARGUE_ARGUMENTS_FILE" <<'THOTH_ARGUE_ARGUMENTS_EOF'
$ARGUMENTS
THOTH_ARGUE_ARGUMENTS_EOF
"${CLAUDE_PLUGIN_ROOT}/scripts/thoth-claude-command.sh" argue --host claude --thoth-arguments-file "$THOTH_ARGUE_ARGUMENTS_FILE"

Response Contract

Treat the structured bridge payload above as the only authority for this command invocation.
If bridge_success is false, report the exact bridge failure and stop.
If run or loop is missing --work-id, show returned candidates and stop; do not invent, create, compile, or guess a work item.
If bridge_success is true and runtime events are present, summarize progress, terminal status, and risk from those events only.
Do not hand-edit .thoth or manually call runtime protocol commands; the Thoth RuntimeDriver advances phases.
If packet.dispatch_mode is external_worker, do not duplicate the work locally; report the run id, worker mode, and the correct follow-up only.
If you only describe what should happen next instead of reporting the executed runtime result, treat that as failure.
For argue, use the returned argument artifacts as evidence; do not replace attacker/adjudicator output with a local summary.
If target resolution is ambiguous, use AskUserQuestion to ask the user to choose one candidate and stop.
If body.authority_patch_preview.confirmation_required is true, ask for explicit confirmation before running the apply command.

Authority Summary

Route

route_class: live_intelligent
intelligence_tier: high
packet_authority_mode: argument_record

Objective

Run a human-quality adversarial discussion: resolve the intended target, ask if ambiguous, let an attacker challenge the direction from first principles, then let an independent adjudicator decide decision_impact and preview any authority patch.

Read the full file on GitHub · 67 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 67 lines · 25 tokens per session scan A d7c1f384fd74

Subscribe to this mod's changes

thoth:argue is a command published in the GitHub repository SeeleAI/Thoth (51 stars, last pushed 23d ago), licensed MIT. It adds 25 tokens to every session and 788 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Orchestration Planning autoresearch claude-code codex dashboard harness

thoth:argue

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

/thoth:argue

Generated Surface

Real Runtime Execution

Response Contract

Authority Summary

Route

Objective

What this file has done since we first saw it

Other commands, from other repositories

build

project

send

qabot-report

ci-report

autobot-list