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 zhaixin244-wq/fnw --skill debategit clone --depth 1 https://github.com/zhaixin244-wq/fnwWrote 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/zhaixin244-wq/fnw/debate)<a href="https://agentmods.dev/skills/zhaixin244-wq/fnw/debate"><img src="https://agentmods.dev/badge/skills/zhaixin244-wq/fnw/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/zhaixin244-wq/fnw/debate"><img src="https://agentmods.dev/badge/skills/zhaixin244-wq/fnw/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 | $0.00092 | $0.06620 |
| Opus 5 | $0.00046 | $0.03310 |
| Sonnet 5 | $0.00018 | $0.01324 |
| Haiku 4.5 | $0.00009 | $0.00662 |
Grade C, and why
debate scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
Wait for user confirmation. If the user declines, clean up (`rm -rf "$DEBATE_DIR"`) and stop. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
/debate --provider glm - GLM (Zhipu AI, glm-5.1) via curl How it starts
The opening of the file, as written. The whole thing — 591 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/debate - Cross-Model Adversarial Review
Invoke an external LLM to adversarially challenge a plan, bug diagnosis, or codebase. You (Claude) are the orchestrator - the reviewer is a different model running as a read-only subprocess. Cross-model diversity is the point: different model families catch different blind spots.
Invocation
/debate - plan review (default), uses first available provider
/debate debug - debugging review mode
/debate review - code review (recent changes via git diff)
/debate review src/lib/ - code review of specific path
/debate --provider codex - Codex CLI / ChatGPT (recommended for diversity)
/debate --provider gemini - Gemini CLI
/debate --provider claude - Claude CLI (same-family fallback)
/debate --provider kimi - Kimi CLI (Moonshot AI, kimi-k2)
/debate --provider glm - GLM (Zhipu AI, glm-5.1) via curl
/debate --provider mimo - MiMo (Xiaomi, mimo-v2.5-pro) via curl
Provider selection order (when no --provider flag):
codex- if Codex CLI is installed (cross-model, recommended)gemini- if Gemini CLI is installed (cross-model)kimi- if Kimi CLI is installed (cross-model, Chinese LLM)glm- ifZHIPUAI_API_KEYenv var is set (cross-model, Chinese LLM)mimo- ifMIMO_API_KEYenv var is set (cross-model, Chinese LLM)claude- always available as fallback (same-family, less diverse)
To auto-detect, check which CLI binaries exist: command -v codex, command -v gemini, command -v kimi. For curl-based providers (glm, mimo), check env vars: ZHIPUAI_API_KEY, MIMO_API_KEY. Use the first one found in the order above.
Instructions
When the user invokes /debate, follow these steps exactly.
Step 0: Parse Arguments
Parse the invocation for:
- Mode:
reviewkeyword -> code review mode.debugkeyword -> debug mode. Otherwise -> plan mode. - Review target (review mode only): Any path after
review(e.g.,src/lib/inngest/). If none, use recent git diff. - Provider:
--provider codex,--provider gemini,--provider kimi,--provider glm,--provider mimo, or--provider claude. Default: auto-detect (first available in order: codex -> gemini -> kimi -> glm -> mimo -> claude).
What ships with it
1 file 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.
- 9d ago First seen · 591 lines · 92 tokens per session scan C 1320a6bd8158
debate is a skill published in the GitHub repository zhaixin244-wq/fnw (29 stars, last pushed 3mo ago), licensed MIT. It adds 92 tokens to every session and 6,620 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…