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.
npx skills add yogsoth-ai/stress-test --skill adversarial-debate-truthseekinggit clone --depth 1 https://github.com/yogsoth-ai/stress-testWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/yogsoth-ai/stress-test/adversarial-debate-truthseeking)<a href="https://agentmods.dev/skills/yogsoth-ai/stress-test/adversarial-debate-truthseeking"><img src="https://agentmods.dev/badge/skills/yogsoth-ai/stress-test/adversarial-debate-truthseeking/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/yogsoth-ai/stress-test/adversarial-debate-truthseeking"><img src="https://agentmods.dev/badge/skills/yogsoth-ai/stress-test/adversarial-debate-truthseeking.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00099 | $0.01251 |
| Opus 5 | $0.00049 | $0.00626 |
| Sonnet 5 | $0.00020 | $0.00250 |
| Haiku 4.5 | $0.00010 | $0.00125 |
Grade A, and why
adversarial-debate-truthseeking 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 10d 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.
This is a copy
91% identical to adversarial-debate-truthseeking — 19 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Adversarial Debate (Truth-Seeking Variant)
A retuning of classic critic-defender-judge debate. The classic version asks "which side argued better?" and outputs a survival/resilience verdict — a persuasiveness metric. That is wrong for research: a claim can win a debate by being slippery (un-pin-down-able) while being scientifically empty. This variant changes the roles and the judge's job so the debate produces a falsifiability classification, not a winner.
What changed from the original (multiagent-debate)
| Element | Original (publication) | This variant (truth-seeking) |
|---|---|---|
| Defender's job | Make the claim look strong / survive | State the claim in its MOST falsifiable form — maximize what it forbids |
| Critic's job | Find weaknesses to score against | Find one concrete observation/computation that would refute it |
| Judge's verdict | Winner + resilience score | Bucket: BROKEN / CORROBORATED / UNFALSIFIABLE |
| Success | Artifact survives | We learn whether the claim is even testable, and if so whether it holds |
| Slippery claim | Wins (un-attackable) | Flagged UNFALSIFIABLE (worst outcome) |
Roles
Defender (steelman-to-falsifiable)
The defender does NOT defend the claim as comfortable or vague. The defender's sole job is to restate the claim in the sharpest, most-forbidding form that is still faithful to what we actually meant. A claim of the form "the unification is elegant" is not a defendable form — it forbids nothing. "Under intervention X the law collapses to form A and NOT form B" is — it stakes out something an observation could contradict. If the defender cannot produce a forbidding form, that itself is the verdict (UNFALSIFIABLE) — the defender must report this honestly rather than retreat to vagueness.
Critic (refuter)
The critic does NOT accumulate debating points. The critic tries to produce ONE of: (a) a counterexample (a case the claim forbids but that occurs / could occur), (b) a derivation error (the claim does not follow from its stated premises), (c) a demonstration that the claim's "forbidden" set is empty (it forbids nothing → UNFALSIFIABLE). The critic must commit to a specific refuter before arguing, to prevent goalpost-shifting.
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.
- 10d ago First seen · 62 lines · 99 tokens per session scan A abcdac60b2b0
adversarial-debate-truthseeking is a skill published in the GitHub repository yogsoth-ai/stress-test (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 99 tokens to every session and 1,251 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. It is 91% identical to adversarial-debate-truthseeking, differing in 19 lines, and is treated as a copy.
Other skills, from other repositories
tournament-judge
Runs a domain-neutral, evidence-based tournament in which independent candidates are reviewed, anonymized, and scored by a blind judge against a rubric fixed before judging. Use when comparing competing plans, designs, implementations, documents, tools, vendors, strategies, or other substantial options; when the user…
sector-rotation
An analysis framework for comparing industries in the Chinese A-share stock market, using business conditions, price momentum, valuation, and money flows. It produces rankings and higher- or lower-allocation suggestions.
oma-scholar
Scholarly research companion using Knows sidecar spec (.knows.yaml). Generates, validates, reviews, queries, and compares structured research-paper sidecars, and fetches them from knows.academy. Use for academic literature search, survey synthesis, paper authoring assistance, and peer review with token-efficient…
browser_cdp
A health check for Python code that produces a score, letter grade, and measures of issues such as complexity, duplication, dead code, dependencies, and architecture.
shogun-screenshot
A screenshot tool for getting images from a computer or web page and then cropping, resizing, or masking sensitive information. Playwright is a browser-automation tool used here to capture web pages.
creating-skills
Guide for creating Claude Code skills following Anthropic's official best practices. Use when user wants to create a new skill, build a skill, write SKILL.md, update an existing skill, or needs skill creation guidelines. Provides structure, frontmatter fields, naming conventions, and new features like dynamic context…