Qiushi-Skill is a collection of agent skills that turn dialectical materialist and practical philosophy into methods for investigating problems, identifying their central contradiction, testing conclusions, and continuing work. It is intended to guide AI agents through analysis and task execution. The catalogue entries are its skills, commands, hook, plugin, and agent components.
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 HughYau/qiushi-skill --skill criticism-self-criticismgit clone --depth 1 https://github.com/HughYau/qiushi-skillWrote 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/hughyau/qiushi-skill/criticism-self-criticism)<a href="https://agentmods.dev/skills/hughyau/qiushi-skill/criticism-self-criticism"><img src="https://agentmods.dev/badge/skills/hughyau/qiushi-skill/criticism-self-criticism/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/hughyau/qiushi-skill/criticism-self-criticism"><img src="https://agentmods.dev/badge/skills/hughyau/qiushi-skill/criticism-self-criticism.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00197 | $0.00906 |
| Opus 5 | $0.00098 | $0.00453 |
| Sonnet 5 | $0.00039 | $0.00181 |
| Haiku 4.5 | $0.00020 | $0.00091 |
Grade A, and why
criticism-self-criticism 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 4d 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
批评与自我批评
"房子是应该经常打扫的，不打扫就会积满了灰尘。" —— 《论联合政府》
目的是治病救人：每指出一个问题，就给一个改进；肯定做得好的部分；对事不对人。
用 / 不用
用：
- 一项工作完成，交付前审视
- 收到用户或审查者的批评
- 同类错误反复出现，要找根本原因
- 阶段性节点复盘
不用：
- 任务刚开始，没有可审视的成果
- 用户在实时指导，反馈已即时纳入
- 紧急修复进行中：先修完，再复盘
操作规程
自我批评
- 复述原定目标与验收标准。
- 重新看制品本身（diff、输出、测试结果），不看自己的叙述；自己知道的问题必须主动暴露。
- 逐项对照 review-checklist.md：完整性、正确性、方法论、质量。
- 问题写具体：哪一步、哪个文件、造成什么影响、根本原因是哪条方法论失误（跳过了调查？没验证？分散了力量？）。
- 输出审视报告（模板见下）。问题表为空时，必须写"未发现需要改正的问题，原因是 ……"。
- 需要独立视角时，派遣
self-critic子 agent 在新鲜上下文中审查制品；主线不得替它预填结论。
处理外来批评
- 先完整听完，不急于辩解。
- 用事实核对：对 → 立即接受并改正，说明改了什么；不对 → 用证据说明，不情绪化。
- 表面接受实际不改，比拒绝批评更坏。
输出模板
## 工作审视报告
原定目标 / 验收标准：……
完成情况：[x] …… [ ] ……（原因）
| 严重程度 | 问题（具体到步骤 / 文件） | 根本原因 | 改进 |
|---|---|---|---|
| 必须改正 | | | |
| 应当改正 | | | |
| 建议改进 | | | |
做得好的地方：……
下次重点关注：……
纪律
- 禁止"做得不够好"这类无信息量描述。
- 禁止没有改进建议的批评，禁止针对性格的评价。
- 禁止基于猜测的批评；不确定的标"存疑"。
- 报告末尾把"下次重点关注"直接交给用户：这是跨会话唯一可靠的跟进机制。
交接
- 报告中的"必须改正" → 立即修复并回到
practice-cognition验证 - 问题根源是主要矛盾没抓对 →
contradiction-analysis
原著依据：original-texts.md
What ships with it
2 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.
- 4d ago Changed · -120 lines · +94 tokens per session b479d79e3b87
- 12d ago First seen · 194 lines · 103 tokens per session scan A 18fd8c07f534
criticism-self-criticism is a skill published in the GitHub repository HughYau/qiushi-skill (3,773 stars, last pushed 5d ago), licensed MIT. It adds 197 tokens to every session and 906 once invoked, about $0.0010 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 2026-08-30.
Other skills, from other repositories
refactor
Expert code refactoring based on Martin Fowler's catalog — improve maintainability without changing behavior. Covers code smells, composing methods, moving features, organizing data, simplifying conditionals, method calls, and generalization. Triggers on: refactor, 重构, clean up, improve code, code smell, extract…
understand
Review and understand freshly AI-generated code. Scans the current repo's uncommitted/branch changes and builds a Claude-style light-themed two-column webpage — left is the project folder tree of changed files, right is the selected file's syntax-highlighted diff (additions/deletions clearly distinguished from…
fix-pr-review-loop
Use when the user asks to fix a PR review and drive it to approval autonomously — "fix the PR review and loop until approved", "fix-pr-review-loop", "keep addressing review comments until this PR is approved". Takes an optional PR number/URL (defaults to the current branch's PR). Repeatedly calls fix-pr-review, waits…
validate-issue
Use when the user asks to validate, review, or check a GitHub issue against the code. Returns a cited update decision with a complexity score.
fable-validate-loop
Use when the user asks to validate a GitHub issue with Fable 5.1 and then autonomously drive it to a reviewed PR in one shot — "fable-validate-loop", "fable validate and work on.
fix-pr-review
Use when the user asks to fix, address, or respond to a PR review — "fix the PR review", "/fix-pr-review". Optional PR number/URL (defaults to the current branch's PR); optional codex argument selects Codex.