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 mguozhen/multi-platform-publisher --skill self-media-opsgit clone --depth 1 https://github.com/mguozhen/multi-platform-publisherWrote 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/mguozhen/multi-platform-publisher/self-media-ops)<a href="https://agentmods.dev/skills/mguozhen/multi-platform-publisher/self-media-ops"><img src="https://agentmods.dev/badge/skills/mguozhen/multi-platform-publisher/self-media-ops/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/mguozhen/multi-platform-publisher/self-media-ops"><img src="https://agentmods.dev/badge/skills/mguozhen/multi-platform-publisher/self-media-ops.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 | $0.00121 | $0.01932 |
| Opus 5.5 | $0.00048 | $0.00773 |
| Sonnet 5.5 | $0.00024 | $0.00386 |
| Haiku 4.5 | $0.00012 | $0.00193 |
Grade A, and why
self-media-ops scanned grade A with 1 finding 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 22d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- X trend：`curl -A "<iPhone UA>" https://x.com/i/trending/<id>`，og:title/og:description 直接给出主题；trend id 是 snowflake，`(id>>22)+1288834974657` 得毫秒时间戳。 How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
自媒体运营 Skill（Hunter 在跑）
铁律（违反任何一条=返工）
- 只进草稿箱，绝不群发。发布=push 到 draft box，群发永远由 Hunter 手动。
- 标题用「事实钩子式直陈」：仍是新闻直陈句、仍禁悬念/对比/转折，但要把这篇最猛的事实或数字放进标题（「三个人，一个月，100 万美元 ARR」式）。基于某篇文章/视频写作时，标题逐字用原文标题。
- deslop 机检必须通过（scripts/deslop_check.py）。高频踩雷：破折号——≤5 处；黑名单词：闭环、对齐、打法、解锁、体感、中文圈（注意"封闭环境"含"闭环"的子串误报，重写绕开）。
- 文末必有两节：「参考来源」（分级标注：官方原文 / 三方报道交叉 / X 单一来源 / 未核实，如实写"未看完整节目""未核对仓库"）＋「利益相关」（AI 客服按结果计费 + LLM 网关 Flatkey，与主题相关就写）。
- 每张图发前必须 Read 目检。配图语言随正文语言（中文文章配中文图），figs/ 按 zh|en|ja 分目录。
- 文末关注钩子：渲染器已自动加「关注 Agent101」行（主号 2026-09 定为 Agent101），不要删。
- 对自己不利的事实照写（BlockRun 体）：自家产品的短板、实测失败、来源拿不到，都如实交代，这是信用资产。
排版分工（三套复刻器，全部实测参数）
| 场景 | 排版 | 渲染器 | Playbook |
|---|---|---|---|
| 新闻/发布/融资（默认） | 晚点 LatePost：黄荧光系统，金句前置，克制白描长段落 | scripts/build_wechat_latepost.py | playbooks/wechat-latepost-playbook.md |
| 强节奏爆点新闻（指定时用） | 量子位：绿色系统，短段落钩子，一键三连尾块 | scripts/build_wechat_qbit.py | playbooks/wechat-qbit-playbook.md |
| 观点文/组织文 | 深思圈：Optima 字体，编号小节，零装饰 | scripts/build_wechat_sisiquan.py | 参数注释在脚本头部 |
Markdown 约定（晚点）：%% 导语｜==x==黄底引语｜> 引用块｜## 黄底居中H2｜图片下一行*…*=分析型图注｜---后=灰色小字来源。
量子位：**x**=绿色加粗结论句｜## =绿左条H2｜图注用*…*。
标准流水线（一篇文章 ≈ 30-50 分钟）
- 选题输入
- X trend：
curl -A "<iPhone UA>" https://x.com/i/trending/<id>，og:title/og:description 直接给出主题；trend id 是 snowflake，(id>>22)+1288834974657得毫秒时间戳。 - X 搜索素材：OAuth1 签名调
/2/tweets/search/recent（密钥~/.secrets/x.env，注意set -a && source才能导出）。 - YouTube/播客：yt-dlp 或 curl 下 mp3 →
~/Library/Python/3.9/bin/mlx_whisper <f> --model mlx-community/whisper-large-v3-turbo（80 分钟音频约 8-10 分钟）。 - 公众号文章：curl + iPhone UA 可拿全文；正文在
id="js_content"。 - 播客雷达：
~/content-radar/radar.py --days 7 --min 45。
- X trend：
- 事实核对：官方原文优先（WebFetch），三方报道交叉（Engadget/The Register/LawSites 这类），引语必须在转写稿里 grep 到原句才算核实。厂商口径一律标注"自家数字，无第三方复测"。
- 写作：新建
~/<slug>-<style>/（article.md + figs/zh/）。写完跑 deslop，改到过；然后按lieflat/SKILL.md的 11 条白名单规则做去 AI 味 pass（翻案腔/顿号罗列/相邻句同款/破折号/冒号提示语/序数小标题/拟人喻体/概括盖数据/起手式/五种翻译腔/段首零主语），信息守恒：未命中规则的句子逐字保留。 - 配图：新闻用真实截图（headless Chrome
--screenshot官方公告页/报道页，sips 转 1080 jpeg）；观点文用单线编辑插画（gpt-image-2 走router.flatkey.ai/v1/images/generations，key=~/.secrets/flatkey-cc.env的 FLATKEY_CC_KEY；基础 prompt 见 references/editorial-line-prompt.txt；挂了换 gemini-3-pro-image 走 chat）。 - 封面：
scripts/build_cover_songti.py --kicker "硅谷前沿 AI 观察 · Agent101" --title "<标题>" --bg <线稿底图> --layout band（底图=单线画、内容压在下三分之一的横构图）。封面文案必须与最终标题一致。 - 推草稿：改 scripts/push_to_wechat.template.py 顶部 TITLE/DIGEST。DIGEST 当第二标题写（订阅号消息里显示这两行，决定打开率），要给钩子而不是内容概述。→
set -a && source ~/.secrets/wechat.env && set +a && python3 push_to_wechat.py。需要换封面重推时：先推新草稿，再draft/delete旧 media_id。 - 交付：
open封面给 Hunter；汇报只说结论+骨架；提醒"只进草稿箱"。
What ships with it
22 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.
- lieflat/.gitignore 151 B
- lieflat/LICENSE 1.0 KB
- lieflat/README.en.md 22 KB
- lieflat/README.md 19 KB
- lieflat/RESEARCH.en.md 22 KB
- lieflat/RESEARCH.md 18 KB
- lieflat/scripts/check-structure.py 4.2 KB runs code
- lieflat/scripts/check-translationese.py 4.7 KB runs code
- lieflat/scripts/compare-human-ai.py 6.4 KB runs code
- lieflat/SKILL.md 28 KB
- playbooks/wechat-latepost-playbook.md 3.1 KB
- playbooks/wechat-qbit-playbook.md 2.4 KB
- references/article-dir-convention.md 518 B
- references/editorial-line-prompt.txt 738 B
- scripts/build_cover_songti.py 10.0 KB runs code
- scripts/build_cover_v2.py 3.9 KB runs code
- scripts/build_wechat_latepost.py 4.2 KB runs code
- scripts/build_wechat_qbit.py 4.9 KB runs code
- scripts/build_wechat_sisiquan.py 3.6 KB runs code
- scripts/bundle_drafts.py 3.1 KB runs code
- scripts/deslop_check.py 4.1 KB runs code
- scripts/push_to_wechat.template.py 4.9 KB runs code
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.
- 22d ago Changed · +5 lines 8989b3f4e61a
- 27d ago First seen · 46 lines · 121 tokens per session scan A 02d88bca2756
self-media-ops is a skill published in the GitHub repository mguozhen/multi-platform-publisher (10 stars, last pushed 22d ago), licensed MIT. It adds 121 tokens to every session and 1,932 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-20.
Other skills, from other repositories
opik-diagnose
Surface the Opik traces worth a developer's attention, ranked by signal — Diagnostics issues first, then errors, failed tool calls, latency, regressions, and low online-eval scores. With the Opik MCP connected it lists the project's agentinsightsissue entities, offers to turn Diagnostics on when the project has it…
opik-test
Turn a failing Opik trace (or a described failure) into a repeatable regression check — a test-suite item with the trace's input and one or two binary assertions — so a fix can be verified by the compare skill. Works over the SDK; uses the MCP write tool when connected. Returns the suite, the item, and the assertion.…
qmclaw-workbench
A local simulator for superconducting quantum-bit experiments, including measurements used to study and tune a qubit. It produces bounded synthetic results without connecting to laboratory hardware or quantum cloud services.
cortex-explore-memory
Explore the memory system's state, find gaps in knowledge, assess coverage, and get diagnostic information. Use when the user asks 'what does my memory look like', 'show me memory stats', 'what am I missing', 'how good is my knowledge', 'memory health', 'show coverage', 'find gaps', 'what topics are weak', or when you…
quantum-fermi-hubbard-vqe
A variational quantum simulation of the open-boundary Fermi–Hubbard chain, a model of particles hopping between sites and interacting when they occupy the same system. It maps the model to qubits with the Jordan–Wigner transformation and searches for a low-energy state.
quantum-grover
A Qiskit implementation of Grover's algorithm, a quantum search method that increases the chance of finding a marked item. It can explain the method, build an explicit quantum circuit, or run it with chosen inputs.