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 xr843/fojin --skill sutra-commentarygit clone --depth 1 https://github.com/xr843/fojinWrote 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/xr843/fojin/sutra-commentary)<a href="https://agentmods.dev/skills/xr843/fojin/sutra-commentary"><img src="https://agentmods.dev/badge/skills/xr843/fojin/sutra-commentary/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/xr843/fojin/sutra-commentary"><img src="https://agentmods.dev/badge/skills/xr843/fojin/sutra-commentary.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 | $0.00139 | $0.01656 |
| Opus 5 | $0.00069 | $0.00828 |
| Sonnet 5 | $0.00028 | $0.00331 |
| Haiku 4.5 | $0.00014 | $0.00166 |
Grade A, and why
sutra-commentary 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 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.
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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
经注对读 — 历代注家怎么讲这一句
用 fojin 的经注对齐数据,把「同一句经,历代各家怎么讲、分歧在哪」做成一份 每一条都能点回原书的对照。
这不是让你凭记忆谈注疏。凡是本 skill 产出的注文,都必须来自工具返回的原文。
前置:确认工具在
需要 fojin MCP(commentaries / verify_quote / read_passage)。没有就先:
claude mcp add --transport http fojin https://mcp.fojin.ai/mcp
工作流
第一步 — 先确认这句经是真的
如果经文是用户凭记忆写的、或来自二手引用,先 verify_quote。
跳过这一步的后果不是"少一道保险",而是你可能在为一句根本不存在的经文编造注解—— 而伪托佛经的句子恰恰是最常被拿来问的(「一花一世界」「前世五百次回眸」都不在藏经里)。
verbatim: false→ 停下来告诉用户这句话不在藏经中,给出closest作为线索。不要继续找注。verbatim: true但cite_matched: false→ 话是真的、出处是错的。先纠正出处,再往下走。
第二步 — 取注
调 commentaries(quote)。
默认 limit 是 8,而一段常有 50 家。 要做对比就把 limit 提到 20–30;
只要 truncated: true,你的输出里必须写明「本段共 N 家,此处呈现 M 家」。
matched: false 时看 available_sutras:
⚠️ 「这部经还没有数据」不等于「历代无人注过这句」。 说成后者是伪造学术结论。正确说法是:「fojin 目前只覆盖《金刚经》, 这一句所属的经尚未收录,因此无法回答,而不是无人注过。」
第三步 — 读懂三个最容易误读的字段
| 字段 | 它的意思 | 误读会造成什么 |
|---|---|---|
tier A/B/C |
这部书的对齐质检档次(程序把注文对到经文上的准确率) | 误当成"这部注疏水平低"。憨山、吉藏的书也可能是 C 档——那说的是对齐难度,不是他的学问 |
same_as |
同一部书收在另一部藏经里的另一版本 | 把玄宗御注的两个版本当成两家独立意见,凭空制造共识 |
truncated / total |
本段实有 N 家,你只拿到 M 家 | 把样本当全集,得出"只有三家注过"这种错误结论 |
note 是截断过的(截到该注疏下一处牒文,最多四行)。要引长段,循 anchor
回原书,不要把截断处当作注家话说完了。
第四步 — 组织成对读
不要平铺直叙地罗列。按这个顺序:
- 经文本身 —— 用工具返回的
passage,标出base_urn - 各家分组 —— 能判定宗派/时代的按宗派或时代分组(唯识、天台、华严、禅、净;唐、宋、明清)
- 共识 —— 多家一致的地方,点出是哪几家
- 分歧 —— 这是对读的价值所在。哪一家讲得不同、不同在哪
- 存疑 —— 遇到 C 档或置信度低的条目,标出来而不是悄悄用掉
判定宗派要有依据(作者、书名、行文),判不出就说判不出,不要按书名猜。
第五步 — 收尾必须声明边界
结尾固定交代三件事,一句话即可:
- 本段共 N 家注,呈现了 M 家
- 覆盖不完整:一部注疏实际所注约有一半没被对齐出来,所以「某家未出现」只意味着数据里没有,不代表他没注过
- 原文出自 CBETA(CC BY-NC-SA 4.0)
输出样式
## 經文
「應無所住而生其心」 —— 《金剛般若波羅蜜經》 fojin:cbeta/T0235.1
## 各家
**唐 · 御注一系**
- 玄宗《御注金剛般若經》[A]:「知色相空,心無所住,故於無所住,生清淨心……」
(F03n0100 与 ZW10n0081 为同书两版,此处计作一家)
**唐 · 法相一系**
- 窺基門下《金剛般若經贊述》[A]:「此第三示勸也……」
## 分歧
玄宗从「心无所住」立说,重在遣执;《贊述》则依論释科判,重在判位……
## 边界
本段共 50 家注,上列 6 家。覆盖不完整——未出现的注家只表示数据里没有,
不表示他没注过此句。原文出自 CBETA(CC BY-NC-SA 4.0)。
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 · 110 lines · 139 tokens per session scan A f92418cbe8b7
sutra-commentary is a skill published in the GitHub repository xr843/fojin (340 stars, last pushed 5d ago), licensed Apache-2.0. It adds 139 tokens to every session and 1,656 once invoked, about $0.0007 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
compare-masters
A comparison skill that explains how different Buddhist teachers or traditions answer the same question.
master-buddhaghosa
A reference skill for answering questions about Theravāda Buddhism, a Buddhist tradition that preserves teachings in the Pali language, with a focus on the writings of Buddhaghosa, a fifth-century scholar.
master-debate
A skill for running an adversarial, multi-round debate between two Buddhist teachers. It is designed for questions where the teachers argue from different positions.
master-mahasi-sayadaw
A reference skill for answering questions about Mahāsi Sayādaw, a Burmese teacher of insight meditation. It explains noting practice, mindfulness, meditation progress, and dealing with distraction.
master-tsongkhapa
A reference-based assistant for questions about Tsongkhapa and the Gelug school, a Tibetan Buddhist tradition. It uses selected Buddhist texts and can answer in Tsongkhapa’s historical teaching style.
master-ajahn-chah
A reference-based assistant for Theravada Buddhism, an early Buddhist tradition, and the Thai Forest meditation tradition. It can answer in Ajahn Chah’s teaching style using selected teachings and Buddhist scriptures.