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 agentmods add skills/guage1029/ai-write-flow/skillnpx skills add guage1029/ai-write-flow --skill skillgit clone --depth 1 https://github.com/guage1029/ai-write-flowWrote 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/guage1029/ai-write-flow/skill)<a href="https://agentmods.dev/skills/guage1029/ai-write-flow/skill"><img src="https://agentmods.dev/badge/skills/guage1029/ai-write-flow/skill.svg" alt="Measured on agentmods" 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 | $0.00179 | $0.05035 |
| Opus 5 | $0.00089 | $0.02518 |
| Sonnet 5 | $0.00036 | $0.01007 |
| Haiku 4.5 | $0.00018 | $0.00504 |
Grade A, and why
ai-write-flow 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.
How it starts
The opening of the file, as written. The whole thing — 364 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ai-write-flow — 技术博客写作工作流
快捷入口：审校模式
优先级： 高于所有 Step，包括 Step 0。检测到本意图时直接进入 Review 流程，不执行输出模式路由。
触发意图： 用户传入已有文章 + 表达"降AI味 / 审校 / 去掉AI腔 / 润色"等意图
跳过步骤： Step 0 及 Steps 1-4 全部跳过
直接加载： references/fact-policy.md + references/style-guide.md + references/checklist.md + references/ai-flavor-patterns.md + references/human-voice-policy.md
执行流程： 直接进入 Step 5 审校（内容 → 作者手迹保护 → AI 味检测/降噪 → 细节 → 二次扫描）
输出结构： 审校报告（对话窗口展示）+ 关键改动说明 + 修订后全文；若用户要求“只检查/先别改/detect/audit only”，只输出检测报告，不输出修订全文
事实处理规则（审校模式专项）：
- 审校模式不得新增、改写或强化高风险事实（版本号、性能数据、价格、模型名、发布时间等）
- 若原文中发现缺少来源支撑的高风险事实，在审校报告的"内容审校"栏列为「待核验事实」，不替用户补全或修改
- 文风润色不改变事实表述；如要修改数字或版本号，必须先告知用户并等待确认
Step 0：输出模式路由（Output Mode Routing）
触发条件： 工作流启动时自动执行（审校快捷入口优先级更高，命中时跳过本步骤）
路由规则：
| 触发关键词 | 输出模式 | 进入流程 |
|---|---|---|
| 图卡 / 贴图 / 卡片 / 小红书 / card / 图卡生成系统 / 1000字以内的内容 | card_post |
加载 card-post-config.md，进入图卡流程 |
| 配图 / 章节配图 / 分段配图 / SVG 插图 / 给文章加图 | longform_article + illustration_extension |
加载 illustration-config.md，进入章节配图流程 |
| 以上均无 | longform_article |
继续 Step 1-6 长文流程 |
illustration_extension 路由规则：
- 用户只要求「给已有文章配图」→ 跳过 Step 1-4 写作流程，直接进入章节 SVG 配图扩展
- 用户要求「写文章并配图」→ 先完成 Step 1-6，文章稳定后再进入章节 SVG 配图扩展
- 未明确要求配图时，不主动生成配图，不打断写作流程
card_post 模式加载文件：
references/card-post-config.mdreferences/fact-policy.mdreferences/research-config.mdreferences/persona.md（语气参考）references/style-guide.md（语气参考，不强制文章结构）
图卡流程（card_post 模式）
card_post 模式共执行 4 步：Step 1（工作区检查）→ Step 2（调研与核查）→ 卡片草稿 → 落盘输出。
Step 1 和 Step 2 与长文主流程相同（见下方主流程章节），但 Step 2 额外执行 research-config.md 中的 card_post 事实核查补充规则，并在 JSON 中输出 card_safe_claims[] 和 excluded_claims[]。
图卡 Step 3：卡片草稿（Card Draft）
触发条件： Step 2 调研通过后执行
加载文件： references/card-post-config.md（已在 Step 0 加载）+ references/fact-policy.md
行为规范：
- 基于 Step 2 的
card_safe_claims[]，决定信息单元数量，输出【卡片数量决策】 - 等待用户确认卡片数量（可微调）
- 用户确认后，按 card-post-config.md 的输出结构，依序输出：
- 【事实核查摘要】
- 【卡片数量决策】（确认版）
- 【发布配文】
- 【图卡内容排版】（封面图卡 + Card 01…）
- 所有图卡要点必须可追溯到
card_safe_claims[]中can_use_in_card: true的 supported claim excluded_claims[]中的事实禁止出现在发布内容（【发布配文】与【图卡内容排版】）
What ships with it
16 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.
- assets/brief-template.md 2.7 KB
- references/ai-flavor-patterns.md 3.6 KB
- references/card-post-config.md 4.2 KB
- references/checklist.md 6.9 KB
- references/fact-policy.md 4.0 KB
- references/human-voice-policy.md 2.5 KB
- references/illustration-config.md 8.1 KB
- references/image-config.md 283 B
- references/persona.md 3.9 KB
- references/research-config.md 7.6 KB
- references/style-guide.md 6.3 KB
- references/workspace-config.md 1.6 KB
- scripts/check_article.py 16 KB runs code
- scripts/check_card_post.py 11 KB runs code
- scripts/check_illustration_plan.py 9.1 KB runs code
- scripts/validate_research.py 4.8 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.
- 4d ago First seen · 364 lines · 179 tokens per session scan A df08f8d5878d
ai-write-flow is a skill published in the GitHub repository guage1029/ai-write-flow (19 stars, last pushed 23d ago), licensed MIT. It adds 179 tokens to every session and 5,035 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…