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 tranfu-labs/tranfu-skills --skill content-productiongit clone --depth 1 https://github.com/tranfu-labs/tranfu-skillsWrote 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/tranfu-labs/tranfu-skills/content-production)<a href="https://agentmods.dev/skills/tranfu-labs/tranfu-skills/content-production"><img src="https://agentmods.dev/badge/skills/tranfu-labs/tranfu-skills/content-production/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/tranfu-labs/tranfu-skills/content-production"><img src="https://agentmods.dev/badge/skills/tranfu-labs/tranfu-skills/content-production.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 | $0.00163 | $0.03830 |
| Opus 5 | $0.00081 | $0.01915 |
| Sonnet 5 | $0.00033 | $0.00766 |
| Haiku 4.5 | $0.00016 | $0.00383 |
Grade A, and why
content-production 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 11d 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 — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Content Production
把本 Skill 当作唯一状态机和验收总控。具体能力由本地 provider Skill 执行;总控不重写 provider 的领域流程,也不允许 provider 创建第二套 run、门禁或用户确认。
完成定义
一次成功 run 必须交付:
- 2 份平台中立母稿。
- 5 个平台 x A/B = 10 份审校终稿。
- 精确 34 个可兑现标题候选和 5 个自动或人工选择结果。
- 5 套入选平台稿配图及优化记录。
- 1 张标题准确、
1923x818的公众号 PNG 封面。 - 1 份验证为零错误、零警告的公众号 clean HTML,以及浏览器预览。
qa.json、qa.md和人工发布交接单。
终点是本地交付包。禁止登录、创建平台草稿、排期、发布或读取发布后数据。
必读资源
开始新 run 或恢复旧 run 前完整读取:
references/workflow.md:入口、阶段、自动决策和恢复顺序。references/capability-contracts.md:九个 provider 槽位与统一调用协议。references/artifact-contract.md:目录、schema、数量和最终验收。references/editorial-policy.md:事实、A/B、平台、标题和视觉边界。references/platform-profiles.json:五平台画像与数量配置。
核心规则
- 只在
runs/<run-id>/写本次产物。provider 只能写总控授权的目录。 run.json、版本、门禁、失效和恢复点只由总控脚本维护。- 所有 provider 调用都使用
capability-contracts.md的 request/result envelope;不要解析 provider 面向人的普通回复来拼装产物。 - provider 未声明匹配的
content-production-provider:*marker,或契约检查失败时,停在初始化。 - 事实与安全 > 平台规范 > 读者画像 > B 风格 > 表达偏好。
- A/B 共享 claims、证据大纲、模型和参数。A 不读取 B 风格;B 只读取本次快照。
- 十份平台稿必须从对应 A/B 母稿生成,不从公众号稿二次派生。
- 标题只能读取审校稳定的十份终稿;内部数字评分不得成为总控接口。
- 公众号封面是独立发布资产,不进入公众号正文图片发布映射
manifest.json。 07-visual/wechat-cover/cover.json只记录生成、候选、视觉 QA 和源文件血缘;发布副本与压缩血缘由 package 阶段拥有。- illustration 先保存原生
manifest.md;所有正文图片优化后再由 package 阶段写发布映射manifest.json。封面保持 PNG 和1923x818。 - image compression provider 只写 staging candidate;总控仅在 candidate 严格更小时采用,否则逐字节保留原图,并由 package 独占发布扩展名、manifest 和 schema v2 optimization。
- 图片写回五份入选 Markdown 后,最后执行公众号排版。
- 整个 run 目录就是唯一交付包;核心阶段 Markdown、正文配图、公众号封面、HTML 和发布资产保留在各自原目录,不复制第二套汇总文件。
handoff.md先列五个平台发布稿、图片目录、封面和 HTML,再按阶段索引 current Markdown、被采用的 prompts 与 native manifests;失败候选、旧 attempt、临时 checkpoint 和内部控制文件不进入运营索引。- 已批准或自动决定的文件禁止覆盖;修订写
.v002、.v003并失效下游。
入口与运行模式
接受且只接受一个创作入口:
--brief <一句话>:运行选题规划并自动选择主选题。--topic <明确题目>:跳过候选发现,仍执行深度调研。--outline <Markdown 路径>:快照大纲,仍执行事实核验和结构规范化。--outline-text <粘贴的大纲>:与大纲文件入口相同。
默认 --run-mode autonomous。该模式保留 topic、outline、titles、visual、final 五个审计门禁,但由总控根据固定规则写决策文件并立即批准,不暂停询问用户。只有用户明确要求逐阶段确认时才使用 --run-mode reviewed。
What ships with it
60 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.
- .gitignore 22 B
- agents/openai.yaml 438 B
- ASSET_NOTICE.md 1.0 KB
- assets/icon.png 488 B
- assets/icon.svg 387 B
- capabilities.yaml 2.6 KB
- LICENSE 1.0 KB
- LICENSES.md 1.6 KB
- README.md 3.4 KB
- README.zh.md 3.1 KB
- references/artifact-contract.md 23 KB
- references/capability-contracts.md 32 KB
- references/core-audience.md 3.3 KB
- references/editorial-policy.md 5.1 KB
- references/platform-profiles.json 7.9 KB
- references/style-b.md 414 B
- references/visual-cardinality-policy.json 2.1 KB
- references/workflow.md 20 KB
- runs/.gitkeep 1 B
- scripts/aggregate-titles.mjs 8.7 KB runs code
- scripts/assemble-publish-packs.mjs 4.9 KB runs code
- scripts/backend-lease.mjs 9.6 KB runs code
- scripts/backend-runtime.mjs 29 KB runs code
- scripts/check-capabilities.mjs 525 B runs code
- scripts/check-claim-regression.mjs 1.8 KB runs code
- scripts/check-provider-result.mjs 6.8 KB runs code
- scripts/contracts.mjs 125 KB runs code
- scripts/cover-queue.mjs 11 KB runs code
- scripts/create-compression-requests.mjs 5.3 KB runs code
- scripts/create-drafting-request.mjs 25 KB runs code
- scripts/create-illustration-request.mjs 16 KB runs code
- scripts/create-proofreading-request.mjs 9.0 KB runs code
- scripts/create-source-request.mjs 12 KB runs code
- scripts/create-title-request.mjs 9.4 KB runs code
- scripts/create-topic-request.mjs 7.1 KB runs code
- scripts/create-visual-coverage.mjs 6.9 KB runs code
- scripts/create-visual-decision.mjs 2.8 KB runs code
- scripts/create-wechat-cover-request.mjs 10 KB runs code
- scripts/create-wechat-layout-request.mjs 5.0 KB runs code
- scripts/illustration-contracts.mjs 69 KB runs code
- scripts/illustration-queue.mjs 45 KB runs code
- scripts/init-run.mjs 11 KB runs code
- scripts/inspect-run.mjs 6.1 KB runs code
- scripts/lib.mjs 19 KB runs code
- scripts/migrate-v2-to-v3.mjs 8.9 KB runs code
- scripts/package-contracts.mjs 40 KB runs code
- scripts/promote-wechat-layout.mjs 4.4 KB runs code
- scripts/run-image-generation.mjs 5.0 KB runs code
- scripts/scan-path-leaks.mjs 3.0 KB runs code
- scripts/set-gate.mjs 16 KB runs code
- scripts/set-semantic-review.mjs 4.1 KB runs code
- scripts/set-stage.mjs 23 KB runs code
- scripts/set-visual-component.mjs 6.7 KB runs code
- scripts/verify-run.mjs 47 KB runs code
- scripts/visual-cardinality.mjs 42 KB runs code
- scripts/visual-preflight.mjs 8.0 KB runs code
- scripts/visual-reuse.mjs 2.7 KB runs code
- scripts/visual-state.mjs 2.0 KB runs code
- scripts/wechat-cover-contracts.mjs 31 KB runs code
- scripts/wechat-layout-contracts.mjs 38 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.
- 11d ago First seen · 177 lines · 163 tokens per session scan A 3a2782d99e01
content-production is a skill published in the GitHub repository tranfu-labs/tranfu-skills (2 stars, last pushed 2d ago), licensed MIT. It adds 163 tokens to every session and 3,830 once invoked, about $0.0008 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
webgl-holographic-foil
A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.
general-video
Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated…
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
html-ppt-taste-brutalist
16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).
chengfeng-check-updates
An environment manager for a video-editing system. It checks whether its skills and runtime—the software needed to run them—are installed and compatible.
diagnostic-stem-delivery
Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.