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 DENGYUFAN0/guashuai-junshi --skill junshigit clone --depth 1 https://github.com/DENGYUFAN0/guashuai-junshiWrote 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/dengyufan0/guashuai-junshi/junshi)<a href="https://agentmods.dev/skills/dengyufan0/guashuai-junshi/junshi"><img src="https://agentmods.dev/badge/skills/dengyufan0/guashuai-junshi/junshi/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/dengyufan0/guashuai-junshi/junshi"><img src="https://agentmods.dev/badge/skills/dengyufan0/guashuai-junshi/junshi.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 | $0.00188 | $0.00828 |
| Opus 5 | $0.00094 | $0.00414 |
| Sonnet 5 | $0.00038 | $0.00166 |
| Haiku 4.5 | $0.00019 | $0.00083 |
Grade A, and why
junshi 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 12d 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
军师模式（主会话模型主战 · 军师子代理点拨）
本适配器默认搭配：主会话模型（CHEAP）= Sonnet 系列；军师子代理（STRONG）= Opus 系列，对应
agents/advisor.md里的model字段。想换模型，改那一行即可，这份操典不用动。
官方 advisor 模式的本机落地（SWE-bench Pro 约 92% 水平 / 63% 成本，军师平均每任务只被请教约 1 次）。
进入时自检
确认当前主会话正在用 CHEAP 档模型。若仍是 STRONG 档，提醒用户：军师模式的前提是主帅换成 CHEAP 模型（否则省不了额度），请在模型选择器切换后重新输入 /junshi，然后停止。
用兵规则
- 主帅自己打仗。日常执行——写代码、改文件、查资料、跑测试——全部自己（主会话）完成，不请军师。
- 只在关键处请军师。满足以下之一才调用
advisor子代理：- 架构/方案的重大取舍，选错代价高；
- 一个 bug 自己尝试两轮仍无头绪；
- 用户明确要求「问问军师」。
- 庙算后问。请教前先写下自己的倾向方案与备选，请军师裁决与修正，不请军师从零代想——裁决式提问更准也更省。
- 一事一问，粮草随行。每次只问一个聚焦的问题，附上军师下判断所需的最小上下文（相关代码片段、已排除的假设、报错原文）；不把整个任务甩给军师。
- 再三渎，渎则不告（蒙卦）。同一问题不带新证据不得重问；对锦囊不满意，先回战场取新证据再问，不反复用同一批材料追问。
- 锦囊要用。军师给了判断就按它执行；执行中发现锦囊与实际矛盾，带着新证据回去追问（走规则 5），不默默弃用。
- 对用户透明。每次请军师都明说：为什么这一步值得动用贵模型、军师给了什么锦囊。用户能据此校准你请教的频率。
散帐
用户说「散帐」「退出」时结束军师模式，回到普通聊天，并汇总本轮共请了几次军师、各解决了什么。
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.
- 12d ago First seen · 32 lines · 188 tokens per session scan A 396446be8175
junshi is a skill published in the GitHub repository DENGYUFAN0/guashuai-junshi (1 stars, last pushed 2mo ago), licensed MIT. It adds 188 tokens to every session and 828 once invoked, about $0.0009 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
wiki-ingest
Ingest a source into the project wiki as OKF v0.2 markdown. Point at a file, PR, or doc and the wiki-curator extracts knowledge, writes YAML frontmatter, and updates relevant concept pages.
wiki-lint
Health-check the project wiki for OKF v0.2 conformance — missing frontmatter, missing type:, malformed index.md/log.md, stale pages past staleafter, broken cross-references, and coverage gaps.
pixeltamer
Generate, edit, and compose images with gpt-image-2. Two backends — OpenAI API key or codex CLI (uses your ChatGPT subscription, no key needed). Three modes — one-shot generate, multi-image batch with verification, and multi-reference composition (up to 16 inputs blended into one). Use when the user asks to make an…
pandic-office
Convert Markdown to PDF (or DOCX/EPUB/HTML) using the pandoc CLI. Use when asked to produce a PDF report, brief, summary, or any document where the input is Markdown and the output should be a polished, paginated file.
tmux
Manage long-running shell sessions with tmux — start a detached session, run a long task, reattach later, capture output. Use when a task takes longer than a single tool call (build, test, log tail).
web-research
Use this skill for requests related to web research; it provides a structured approach to conducting comprehensive web research.