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 SupSub-AI/supsub-cli --skill supsub-unreadgit clone --depth 1 https://github.com/SupSub-AI/supsub-cliWrote 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/supsub-ai/supsub-cli/supsub-unread)<a href="https://agentmods.dev/skills/supsub-ai/supsub-cli/supsub-unread"><img src="https://agentmods.dev/badge/skills/supsub-ai/supsub-cli/supsub-unread/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/supsub-ai/supsub-cli/supsub-unread"><img src="https://agentmods.dev/badge/skills/supsub-ai/supsub-cli/supsub-unread.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 | $0.00266 | $0.02675 |
| Opus 5 | $0.00133 | $0.01337 |
| Sonnet 5 | $0.00053 | $0.00535 |
| Haiku 4.5 | $0.00027 | $0.00267 |
Grade C, and why
supsub-unread 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- 安装:`curl -fsSL https://raw.githubusercontent.com/SupSub-AI/supsub-cli/master/scripts/install.sh | bash`(native 安装,装到 `~/.local`、支持后台自动更新);或包管理器 `npm i -g @supsub/cli` / `pnpm add -g @supsub/cli` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- 安装:`curl -fsSL https://raw.githubusercontent.com/SupSub-AI/supsub-cli/master/scripts/install.sh | bash`(native 安装,装到 `~/.local`、支持后台自动更新);或包管理器 `npm i -g @supsub/cli` / `pnpm add -g @supsub/cli` How it starts
The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
supsub-unread Skill
未读工作流的入口 skill:回答「我有哪些未读」,并完成「概览 → 钻取 → 整体已读」的闭环。涉及的命令跨 unread / sub / focus 三个命令树,本文档按工作流组织。
Prerequisites
- 安装:
curl -fsSL https://raw.githubusercontent.com/SupSub-AI/supsub-cli/master/scripts/install.sh | bash(native 安装,装到~/.local、支持后台自动更新);或包管理器npm i -g @supsub/cli/pnpm add -g @supsub/cli - 已登录:
supsub auth status显示 Authenticated(首次使用先supsub auth login) - 未授权(exit 2 / UNAUTHORIZED)时不要止步于「你未登录」:直接运行
supsub auth login为用户打开浏览器授权(命令会自动打开浏览器并阻塞等待授权,请用足够长的超时,如 10 分钟;用户只需在浏览器点确认,无需在终端输入任何内容),授权成功后重试原命令。无浏览器 / 无头环境再回退为提示用户SUPSUB_NO_BROWSER=1 supsub auth login。
已读模型:只有「整体已读」,没有「单篇已读」
这是本 skill 最容易踩错的一点,先记住:
- CLI 能做的只有两件事:
sub mark-read --all(这个号我读完了)、focus mark-read --all(这个关注点我看完了)。 - 没有单篇已读。原因是产品侧的:文章在 SupSub 里没有「详情 / 阅读」这一步,也就不存在「用户读完了这一篇」的自然事件,所以 CLI 只保留用户明确表达的整体已读意图。
- 用户说「把这篇标为已读」时:如实说明不支持,再问是否要把整个源 / 整个关注点标为已读。不要试图用
--content-id绕过(会抛INVALID_ARGS),也不要靠循环调用伪造逐篇已读。 - 也没有分组级已读命令。用户要「清空某个分组」时:先
group subs列出组内源,逐个源确认后再逐个sub mark-read --all,每个源都要在确认范围内。
Workflow(三步闭环)
- 概览:
supsub unread -o json→ 一次拿到所有unreadCount > 0的订阅源与关注点。 - 钻取:对用户关心的源/关注点拉未读列表(默认就是未读):
- 订阅源 →
supsub sub contents --source-id <sid> --type <t> -o json - 关注点 →
supsub focus contents --id <fid> -o json
这里用默认(未读)是对的——本 skill 的语境本来就是「我有什么没读」。但若用户中途转成按时间浏览 (「那 X 最近都讲了什么」),要改用
--all并把isRead当标注,详见supsub-sub/supsub-focus。 - 订阅源 →
- 整体已读(仅当用户明确表达「读完了 / 清掉」时):
mark-read --all,必须先二次确认(见下)。
翻页策略(不要反复询问用户)
contents 默认一页 20 条(--page-size 上限 100)。按用户意图选择策略,过程中不逐页询问:
- 浏览类意图(「看看有什么未读」):只拉第 1 页,结合
unreadCount告知「共 N 条未读,已显示前 20 条」;用户说「继续 / 看更多」再--page 2。 - 全量类意图(「全部未读」「统计 / 导出」):自动循环翻页——用
--page-size 100减少请求数,返回条数 < page-size 即最后一页;unreadCount可预估页数。这类批量拉取一律加--brief(只留 contentId/title/时间/已读状态,100 条约 14KB;不加约 90KB,会撑爆工具输出上限被迫落盘)。 - 注意:「清未读」不需要先把文章全部拉下来——整体已读是一条命令的事,拉全量只在用户要看 / 要统计时才做。
--all 二次确认(强制规则)
执行 sub mark-read --all 或 focus mark-read --all 前,必须依次完成:
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 · 137 lines · 266 tokens per session scan C 9bdf4b391181
supsub-unread is a skill published in the GitHub repository SupSub-AI/supsub-cli (5 stars, last pushed 1mo ago), licensed MIT. It adds 266 tokens to every session and 2,675 once invoked, about $0.0013 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
hermes
Send Telegram messages from an agent through the Hermes MCP telegram tool, with the Hermes CLI (@andrebuilds/hermes) as a fallback. Use when a user asks to send a Telegram message, mentions Hermes, wants to interact with the Hermes toolset, asks to verify Hermes manually, or needs to choose between the Hermes MCP and…
anywrite
Use this skill whenever the user mentions "anywrite" by name, or mentions Anytype, asks to create/update/search/organize notes, tasks, or PKM objects in Anytype, or wants to upload files, manage properties/tags/types, or chat inside an Anytype space. Covers all 52 endpoints of the Anytype local API (spaces, objects…
trce-share
Share the current Claude Code session with trce after an explicit user request.
gws-gmail
Gmail: Send, read, and manage email.
compose:ask
Use whenever you need a decision, clarification, or approval from the user — covers how to ask with the question tool, and how to resolve the decision yourself when no user is available (question tool absent, or a [Never-Ask] response).
recipe-save-email-attachments
Find Gmail messages with attachments and save them to a Google Drive folder.