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.
git clone --depth 1 https://github.com/Xiaoyun-0922/openmarxWrote 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/commands/xiaoyun-0922/openmarx/marx-contradiction)<a href="https://agentmods.dev/commands/xiaoyun-0922/openmarx/marx-contradiction"><img src="https://agentmods.dev/badge/commands/xiaoyun-0922/openmarx/marx-contradiction/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/commands/xiaoyun-0922/openmarx/marx-contradiction"><img src="https://agentmods.dev/badge/commands/xiaoyun-0922/openmarx/marx-contradiction.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 | $0.00026 | $0.00537 |
| Opus 5 | $0.00013 | $0.00269 |
| Sonnet 5 | $0.00005 | $0.00107 |
| Haiku 4.5 | $0.00003 | $0.00054 |
Grade A, and why
marx-contradiction 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.
What it actually says
Use the marx skill in contradiction mode.
Response language policy:
- Match the user's language.
- Keep required contract labels exactly:
已检查证据:,未检查项目证据:,实践方案, and验证标准. - Provide bilingual output only when explicitly requested.
Evidence pass:
- Inspect available project evidence first for concrete project/repository/GitHub/adoption questions.
- Start with
已检查证据or未检查项目证据. - First visible output line must be either
已检查证据:or未检查项目证据:before any title, table, horizontal rule, or summary. Use an ASCII colon:, not a full-width Chinese colon：. Do not use variants such as已检查项目证据:. - Do not start with a horizontal rule, generic title, or
证据表; the evidence line comes first. - Identify one primary contradiction bound to concrete project facts.
- Do not lead with abstract macro-theory.
Evidence freshness:
- Mark prior-turn facts as reused evidence, not freshly inspected evidence.
- If a concrete implementation target matters, inspect it this turn or label it as an assumption.
Compact output:
- Default to a concise answer; avoid large tables unless the user asks for a broad audit or OSS adoption evidence.
- Focus on one contradiction, its major aspect, the immediate lever,
实践方案, and验证标准.
Forbidden output terms:
- Do not mention 使用价值, 交换价值, 社会必要劳动, 劳动价值, 资本论, 生产价值, or 劳动者 unless the user explicitly asks for theory.
- If tempted to use those terms, translate them into concrete project language: target user, visible problem, adoption surface, trust signal, time-to-first-value, distribution channel.
OSS evidence table:
- For stars/users/adoption questions, include a compact evidence table with README positioning, installation friction, demo or screenshot, quickstart, GitHub topics, trust signals, competitive context, distribution evidence, and user feedback loop.
Required final sections:
- Include
实践方案and验证标准exactly in every project-grounded answer. - Do not rename these sections to "next steps", "建议", "接下来的路", or "最小下一步验证".
验证标准must contain an observable pass/fail rule or measurable metric.
User request: $ARGUMENTS
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 · 43 lines · 26 tokens per session scan A 47a046dc2df8
marx-contradiction is a command published in the GitHub repository Xiaoyun-0922/openmarx (5 stars, last pushed 4mo ago), licensed MIT. It adds 26 tokens to every session and 537 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 2026-08-31.
Other commands, from other repositories
workflow-foundation-sprint
Run the Foundation Sprint workflow (2-day strategic-alignment arc producing a Founding Hypothesis).
epic-dev
Epic/sprint development workflow with git worktrees, GitHub issues, and phased delivery.
estimate-work
Estimate a task or project honestly, by decomposing it, surfacing unknowns, and giving a range not a false point.
okr-draft
Draft OKRs with an inspiring objective and measurable, outcome-based key results, avoiding vanity and task-lists.
product-requirements-doc
Write a product requirements document (PRD) that defines the problem, the users, and success, without prescribing the build.
extract-action-items
Pull every commitment, task, and deadline out of meeting notes or a thread, with owners.