TypingMindTypingMindThe best frontend for AI
Blog 3p Harness logo

Blog 3p Harness

Organization
flaqai
blog-3p-harness

初始化、冻结、恢复并机械校验仅人工原生发布交接的博客 3P 内容项目；要求用户先确认完整写前方案和每篇平台/账号映射，严格限制用户提供的平台范围，且不自动发布。用于启动内容项目、准备工作区、汇报写前方案、锁定需求或校验工作流结构。

Overview

Publisherflaqai
Repositorybacklink_skills
Skill nameblog-3p-harness
Stars
760
Forks
237
Bundled files
5
LicenseMIT
Links
Skill sourceRepositoryHomepage
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by flaqai on GitHub. Read the source before you install it.

Installation

Install the Blog 3p Harness AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/flaqai/backlink_skills/tree/main/blog-3p-skill-matrix/skills/blog-3p-harness
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/flaqai/backlink_skills.git /tmp/backlink_skills
mkdir -p .claude/skills
cp -r /tmp/backlink_skills/blog-3p-skill-matrix/skills/blog-3p-harness .claude/skills/blog-3p-harness
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Blog 3p Harness in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Blog 3p Harness on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Blog 3p Harness is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

博客 3P：本地流程工具

先阅读术语约定。本技能负责可恢复的本地工作区和机械校验，不代替编辑判断、平台发布或人工原生编辑器操作。

初始化

运行 scripts/harnessctl.py init --workspace <absolute-path> --campaign-id <id>。它只创建本地文件：campaign.json、state.json、confirmation.md、requirements-contract.md、唯一可编辑的 prewrite-plan.json、确定性只读视图 prewrite-plan.md、一次性 pre-clearance-checklist.md 及标准证据/输出目录。当前 schema-2.14 工作区使用写前方案 schema 1.8、evidence pack schema 1.3 和 article-package.json@1.5，默认持续主会话及 articles/<article_id>/{context,research,canonical,reviews,handoff} 路径隔离，并保存原始用户确认回执；schema-2.13 的独立 R 及更早版本按各自历史规则读取/校验，不自动转成 WQ。当前工作区只支持人工原生发布交接；脚本绝不执行平台写入。

写前方案与用户确认

G 的第一阶段是 prewrite_planning。在用户确认前，不得存在文章隔离路径、W/R 派发、文章队列、大纲、正文、图片、唯一基准稿或文章调研子角色。G 编辑 prewrite-plan.json 后运行 sync-prewrite-plan，再向用户展示生成的 prewrite-plan.md；Markdown 是只读视图，禁止双份手填。

每篇方案卡固定为四个可读区块：任务/受众；独立读者价值和必保留次要 CTA；editorial_brief；风险与待用户决定事项。还须展示 frozen_delivery_mapping、topic_slot（读者任务、核心意图、市场、差异化角度、禁止偏离项）和 evidence_posture。topic_slot 不锁字面关键词/标题。editorial_brief 汇总关键词/本地化路径、事实边界、暂定结构、图文叙事与人工传递假设，不双份维护研究。只展示实际共享证据。冻结 review_effort：默认 AUTHOR_QA_INTEGRATED / WQ_SHARED_CONTEXT / INTEGRATED_IN_AUTHOR_QA；用户要求或具体风险时用 INDEPENDENT_R_ESCALATION / WR_INDEPENDENT，仅 WR 内确有成文前风险才追加独立研究审。不可证实的内容标 UNVERIFIED 并说明风险；写前方案不等于研究批准，W 仍要一体化研究。

只有明确的 OWNER_PREWRITE_PLAN_CONFIRMED 才允许启动文章任务。当前 schema 2.14 须先保存用户确认原文到 evidence/owner-confirmations/，再运行 confirm-prewrite-plan --confirmation-id ... --receipt-file ... --receipt-type OWNER_MESSAGE|OWNER_FILE --source-locator ...。该命令绑定文章集合、计划/回执哈希、受保护范围及来源/时间；手工改状态不能提升方案。语言、市场、平台、账号、适配模式、跨语言例外、topic_slot 或写作姿态变化时，sync-prewrite-plan 返回 SCOPE_RECONFIRM_REQUIRED 且保留旧回执；须显式失效后取得新确认。OWNER_PREWRITE_PLAN_CHANGES_REQUESTED 不允许部分启动。

平台范围与本地化

  • 没有平台映射时，G 只可创建或恢复一个可见、可复用的 blog-3p-platform-matching 子角色；它只在用户提供的 XLSX/表格/消息候选中做语言/市场/内容格式匹配。G 只核对来源与唯一性，不排序、不选平台。
  • 当前 schema 2.14 只使用 HUMAN_RELEASE_ONLY_V1 与 release_policy.mode = HUMAN_NATIVE_ONLY；machine_external_writes_allowed = false。human_release_requested 不能作为跳过人工发布映射的开关。platform_scope.allowed_pairs 仅投影已确认的用户平台选择，每篇在 WQ/WR 派发前须有不同 {platform, account} 与兼容主读者记录；pair 的用户来源哈希、准确定位、平台原文、账号确认原文和确认 ID 缺一不可。
  • 登录页、注册公告、编辑器行为、会话恢复、内部配置和历史记录只能说明可承接性，绝不能新增允许的平台/账号。模糊的“新增语言”或“平台不重复”请求没有平台选择授权，只能返回 OWNER_DECISION_REQUIRED。
  • 文章语言来自用户确认。每篇 locale_platform_validation 必须分开记录 primary_reader_languages、primary_reader_markets、primary_audience_evidence_path 与 transport_supported_content_languages、transport_evidence_path。平台支持语言只决定传递资格，不能证明英语或任何语言是平台主读者语言；默认只有 PRIMARY_AUDIENCE_MATCH 可派发。语言/市场不匹配只能是逐篇、逐字绑定的 CROSS_LANGUAGE_EXCEPTION_OWNER_CONFIRMED，否则写 PLATFORM_AUDIENCE_MISMATCH_RECONFIRM_OWNER，不得自动翻译或改配。
  • 跨语言用语顺序固定为 CURRENT_BRAND_SITE → REGIONAL_SERP → MODEL_TRANSLATION_FALLBACK。Google Trends 如可用，只比较英文种子和英文候选，并且只是全球相对关注度背景；数据不足、无 Breakout 或无清晰趋势时，不得强求或阻塞，继续以当前品牌站与地区 SERP 的有边界证据选择长尾意图。两次地区 SERP 核验都没有可用共识时，才以有据模型翻译兜底。平台资料只能决定表达形式、技术深度和传递风险，不能作为关键词、自然变体、搜索意图、当地需求或热门度依据。

运行与交接

  1. G 完成写前方案、保存并绑定用户确认回执后，运行 check 和 dispatch-readiness。前者仅证明结构可读；后者必须显示 DISPATCH_READY，否则先补齐明确列出的 REQ-*、可见项目 G 登记、每篇人工发布映射或语言/平台兼容记录，再创建 articles/<article_id>/ 的隔离交付路径并派发可见 W/R。build-article-context 的 --workspace 始终是含 campaign.json、state.json、prewrite-plan.json 与 requirements-contract.md 的内容项目根目录；文章产物由其下的确定性 articles/<article_id>/ 路径承接，不能使用新建的空目录作为内容项目根。
  2. blog-3p-writer 是唯一可执行 W；blog-writer-merged 只作编辑参考。W 连续完成研究、成稿、图片和双载荷，再以 CANDIDATE_LOCK → ADVERSARIAL_SELF_QA → TARGETED_REPAIR → FINAL_BIND 自查本篇证据、正文、CTA、图文语义与载荷并绑定最终哈希。默认 WQ 只可登记 AUTHOR_QA_READY，不生成独立 R 的 APPROVED。用户要求或记录高风险时走 WR，独立 R 一次 FULL_REVIEW，仅成文前研究风险另加研究审。机械投影校验正常只语义读 HTML，COMPANION_DUAL_READ_REQUIRED 才双读 Markdown。
  3. blog-3p-gate 在一个 BATCH_GATE_ACCEPTANCE 回合中验收当前所有具备适用质量凭据的文章：WQ 是 W 的可见 AUTHOR_QA_READY 哈希回执，WR 是独立 R 的 APPROVED；升级 WQ 必须有后者。每篇保留独立结论行。W 在质量核验前编译交付页；blog-3p-human-handoff 仅在 HUMAN_RELEASE_READY 后生成派生清单，不改写已核验载荷。
  4. 人工回传 HUMAN_ACCEPTED URL 后，校验回执并生成只读快照；复用同一项目统筹 G 的 PUBLIC_QA_BATCH_READONLY 回合比对当前已回传文章，不重启 W/R。

可见角色会话、文章路径隔离与 Git 工作树是不同概念。默认持续主会话和文章路径隔离：可见 W 可跨普通 WQ 文章复用；WR 才激活独立于 W 的 R。每篇只读写自己的 articles/<article_id>/；全项目只登记一个 G。Git 工作树只在记录真实并发写入、高风险重写/回滚或用户要求 Git 隔离时允许，槽位或 token 节省不是理由。CLI 只可执行确定性本地操作，不能冒充可见 WQ/WR 质量任务。

每次交接前运行 scripts/harnessctl.py check --workspace <campaign-root>。在 WQ 最终自查或 WR FULL_REVIEW 前，W 以内容项目根运行 check-review-ready --workspace <campaign-root> --article-contract articles/<article_id>/context/article-contract.json --review-index articles/<article_id>/reviews/review-index.json --article-package articles/<article_id>/article-package.json。它只校验身份、哈希、来源引用结构、固定载荷、冻结 CTA 与伴随投影，不作质量批准。TOPIC_EVIDENCE_CONFLICT 须同一 G 的有据决策后才能恢复。当前 schema-2.14 再运行 check-article-package、check-handoff-manifest、G 批量 check-batch-gate；它们验证适用质量凭据：WQ AUTHOR_QA_READY、WR 独立 APPROVED，不评分文章。WQ 改稿/改图/改载荷重跑同一 W 的完整自查并换新回执，不走 R-Δ；WR 的有限修复才可由同一 R 作有边界 R-Δ。公开回传先校验 receipt。

需要量化而不是猜测时，运行 summarize-efficiency --workspace <campaign-workspace> [--output evidence/operations/efficiency-summary.json]。它只读取已有任务记录；input_tokens、output_tokens、wall_time_seconds 和 tool_calls 都是可选运行时字段，缺失时明确输出 UNAVAILABLE，不写入文章包、不成为发布门，也不启动模型或新角色。

恢复

发生压缩、恢复或上下文不确定时，先重读流程核心，再读取 campaign.json、state.json、requirements-contract.md、当前文章的 articles/<article_id>/context/article-contract.json 与 articles/<article_id>/reviews/review-index.json，以及当前变更文件。只有哈希漂移、问题项谱系不清、代理替换或升级时，才完整重读历史报告；在状态历史中追加 CONTEXT_REHYDRATED，不得推断缺失证据。

日常先以流程核心和当前结构化记录行动；仅在范围冻结、恢复异常或对应风险触发时，按需读取内容项目契约或写前检查清单的相关段落。机器字段、状态码、命令参数与路径保持不变。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Blog 3p Harness AI skill do?

初始化、冻结、恢复并机械校验仅人工原生发布交接的博客 3P 内容项目；要求用户先确认完整写前方案和每篇平台/账号映射，严格限制用户提供的平台范围，且不自动发布。用于启动内容项目、准备工作区、汇报写前方案、锁定需求或校验工作流结构。

Why use Blog 3p Harness on TypingMind?

Because you install it once and use it with any model. Blog 3p Harness is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Blog 3p Harness in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/flaqai/backlink_skills/tree/main/blog-3p-skill-matrix/skills/blog-3p-harness. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Blog 3p Harness?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Blog 3p Harness?

As many as you like. As long as a model supports skills, you can use Blog 3p Harness with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Blog 3p Harness AI skill free?

Yes. It is published on GitHub by flaqai under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

Related AI Skills

View all
openclaw logo

Block Kit

openclaw
OrganizationPopular

Use proactively for structured or interactive Slack replies, and when asked to author or validate native Slack Block Kit JSON.

82.2K
391K
2 files
View
openclaw logo

1password

openclaw
OrganizationPopular

Set up and use 1Password CLI for sign-in, desktop integration, and reading or injecting secrets.

82.2K
391K
2 files
View
openclaw logo

Apple Reminders

openclaw
OrganizationPopular

List, add, edit, complete, or delete Apple Reminders and reminder lists via remindctl.

82.2K
391K
Instructions
View
openclaw logo

Clawhub

openclaw
OrganizationPopular

Search ClawHub for skills when a requested capability is not already available; install, verify, update, uninstall, publish, or sync skills.

82.2K
391K
Instructions
View

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇