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 Magnus-Gille/claude-skills --skill debate-claudegit clone --depth 1 https://github.com/Magnus-Gille/claude-skillsWrote 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/magnus-gille/claude-skills/debate-claude)<a href="https://agentmods.dev/skills/magnus-gille/claude-skills/debate-claude"><img src="https://agentmods.dev/badge/skills/magnus-gille/claude-skills/debate-claude/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/magnus-gille/claude-skills/debate-claude"><img src="https://agentmods.dev/badge/skills/magnus-gille/claude-skills/debate-claude.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 | $0.00036 | $0.00677 |
| Opus 5 | $0.00018 | $0.00338 |
| Sonnet 5 | $0.00007 | $0.00135 |
| Haiku 4.5 | $0.00004 | $0.00068 |
Grade A, and why
debate-claude 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debate with Claude
This is a Codex-conductor workflow: Codex writes the draft and self-review, invokes Claude as the independent reviewer, and writes the response, summary, and index. Do not have Claude invoke this skill or present a Claude-run self-review as independent cross-model review. Other harnesses may read these references for context, but should route execution back to the Codex conductor.
Use this skill when a consequential architecture, protocol, security assessment, research claim, product priority, or similarly substantive decision benefits from an independent adversarial review. Do not invoke it for typos, formatting, routine implementation details, or mechanical refactors.
Codex owns the artifact, context, model choice, and final decision. This skill does not authorize publication, deployment, credential access, or external messages.
Read only the references needed for the current stage:
- Protocol — the two-round lifecycle, frozen inputs, draft, self-review, response, and rebuttal.
- Reviewer safety — the isolated CLI invocation, model selection, prompt envelope, and output metadata.
- Type lenses — universal and topic-specific self-review and critique questions.
- Artifacts and records — scratch/trackable files, critique-log schema, sanitization, summary, index, and completeness checks.
Non-negotiable boundaries
- Freeze a scoped, sanitized input package before each round. Feed that package through stdin; do not give the reviewer live repository, Munin, filesystem, browser, hook, plugin, or MCP access by default.
- If extra project context is warranted, the conductor selects, sanitizes, and freezes it explicitly in the round package. Round 2 gets a refreshed frozen package that includes the conductor's response and any explicitly selected new context.
- Use the exact model the user specifies. Do not add an automatic fallback or substitute another model. If no model is specified, choose a suitable configured model and record both the requested and actual model from invocation metadata.
- Verify current CLI flags with
claude --helpbefore changing the invocation. The safe baseline is print mode, plan permissions, no session persistence, safe mode, restricted mode, an empty tool list, and an empty strict MCP configuration. Use only flags confirmed by local help. - Record the actual model from structured CLI metadata when available; never treat a model name written in the reviewer's prose as verification.
- Capture stdout yourself and write the debate artifacts. Do not ask Claude to write files or invoke tools.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago Changed · -365 lines · -22 tokens per session cffea881bb17
- 12d ago First seen · 400 lines · 58 tokens per session scan A 17776f8aefc3
debate-claude is a skill published in the GitHub repository Magnus-Gille/claude-skills (5 stars, last pushed 6d ago), licensed MIT. It adds 36 tokens to every session and 677 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…