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 justnau1020/claude-os --skill debategit clone --depth 1 https://github.com/justnau1020/claude-osWrote 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/justnau1020/claude-os/debate)<a href="https://agentmods.dev/skills/justnau1020/claude-os/debate"><img src="https://agentmods.dev/badge/skills/justnau1020/claude-os/debate/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/justnau1020/claude-os/debate"><img src="https://agentmods.dev/badge/skills/justnau1020/claude-os/debate.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 | $0.00043 | $0.01260 |
| Opus 5 | $0.00022 | $0.00630 |
| Sonnet 5 | $0.00009 | $0.00252 |
| Haiku 4.5 | $0.00004 | $0.00126 |
Grade A, and why
debate 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.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debate
Adversarial evaluation of the current state of a feature, system, or codebase area. Five agents argue from different angles, a moderator synthesizes a verdict. The output is a readiness report -- not a plan, not code.
Use this when you don't know if what you have is good enough.
Invocation
/debate [topic] [--scope "focus areas"] [--constraints "known constraints"]
Examples:
/debate "API security posture"/debate "onboarding experience" --scope "auth, setup, first use"/debate "launch readiness" --constraints "feature X is gated behind flag"
Pipeline
Phase 0: Constraints Gathering (Main + User)
Before spawning agents, main MUST:
- Ask the user for known constraints:
- What's in scope for this review?
- Any decisions already made that agents should respect?
- Known limitations or trade-offs to evaluate against?
- Check memory for relevant project constraints
- Compile a constraints block:
CONSTRAINTS (non-negotiable):
- [constraint 1]
- [constraint 2]
DO NOT propose solutions that violate these constraints.
This prevents the #1 failure mode: agents recommending something already ruled out.
Phase 1: Debate (Parallel Agents)
Spawn 5 agents. The 4 debaters run in parallel. Moderator waits for all 4.
| Agent | Role | Method |
|---|---|---|
arch-critic |
Code auditor -- finds gaps, bugs, half-baked features | Reads source code, tests, architecture docs |
comp-critic |
Competitive benchmarker -- holds us to industry bar | Web search for competitors, best practices |
code-defender |
Argues readiness from actual code evidence | Reads source code, test coverage, recent commits |
industry-defender |
Argues competitive positioning from market context | Web search for market context, user expectations |
moderator |
Synthesizes all 4 reports into a verdict | Receives reports via SendMessage, writes report |
Agent prompt template for debaters:
You are [ROLE] in an adversarial debate about: [TOPIC]
Scope: [SCOPE]
CONSTRAINTS:
[constraints block]
Your job: [ATTACK gaps / DEFEND readiness] with EVIDENCE.
- Read the actual code, don't guess
- Cite specific files, functions, line numbers
- Compare against [competitors / best practices]
- Be brutal -- no hedging, no "it depends"
Send your report (max 50 lines) to the moderator agent when done.
Format:
## [Your Role] Report
### Key Findings (numbered, max 10)
### Verdict: [READY / NOT READY / CONDITIONAL]
### Evidence: [file:line citations]
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 · 142 lines · 43 tokens per session scan A 79a56da0654c
debate is a skill published in the GitHub repository justnau1020/claude-os (3 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 43 tokens to every session and 1,260 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
resolve-pr-comments
Evaluate, fix, answer, and reply to GitHub pull request review comments and conversation comments. Handles both change requests (fix or skip) and reviewer questions (explain using reasoning recalled from past Claude Code transcripts). Use when the user asks to "resolve PR comments", "fix review comments", "address PR…
review-code
Review code for bugs, security vulnerabilities, API misuse, consistency issues, simplicity problems, or test coverage gaps by running internal reviews and a peer review in parallel and returning combined findings. Single-concern with a type argument, or full review with no argument. Use when the user asks to "review…
assess-technical-debt
Assess project-wide structural technical debt: complexity hotspots, deprecated API usage, duplication clusters, and architecture rot. Ranks findings by impact and refactor effort into a report at .turbo/technical-debt.md. Use when the user asks to "assess technical debt", "find technical debt", "review technical…
map-codebase
Deep architecture report that fans out parallel inspections across different aspects of the codebase (structure, tech stack, APIs, patterns, data flow, dependencies, testing) and synthesizes findings into a comprehensive document at .turbo/codebase-map.md and .turbo/codebase-map.html. Use when the user asks to "map…
peer-review
Run an independent peer review via Codex. Use when the user asks to "peer review", "peer review my code", "peer review my plan", "get a second opinion", or "independent review".
claude-print
Run a non-interactive Claude Code print-mode call from Codex. Use when the user asks to "claude print", "ask claude", "run claude", "consult claude", or when a Codex Turbo skill needs Claude as an independent peer reviewer.