Outlier Post Finder
ScrapeCreators/social-media-research-skills
A skill your agent uses when the user wants to find posts, videos, reels, shorts, tweets, or social content that overperformed versus a creator, brand, or competitor baseline.
依使用者真實貼文與成效寫 Facebook／Instagram／YouTube／Threads／X 文案，包含 ChatGPT Chat 的「寫文」「Mode C」「用我的格式／口氣」「黑底白字」；本機工作台、規劃、確認後發布、留言回覆及成效學習。使用者說「發文」「文案」「Social Post 介面」「回覆留言」「查流量」「把數據訓練進去」「比較貼文」「優化 pattern」時使用。
Install Social Post
$ npx skills add Hao0321/claude-skill-social-post --skill social-post -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install Hao0321/claude-skill-social-post social-post --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/Hao0321/claude-skill-social-post.git skills-src && mkdir -p .claude/skills && cp -r skills-src/social-post .claude/skills/social-post && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "social-post" agent skill from https://github.com/Hao0321/claude-skill-social-post/tree/main/social-post into .claude/skills/social-post/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "social-post", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/Hao0321/claude-skill-social-post/tree/main/social-postType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add Hao0321/claude-skill-social-post --skill social-post -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install Hao0321/claude-skill-social-post social-post --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Hao0321/claude-skill-social-post.git skills-src && mkdir -p .agents/skills && cp -r skills-src/social-post .agents/skills/social-post && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "social-post" agent skill from https://github.com/Hao0321/claude-skill-social-post/tree/main/social-post into .agents/skills/social-post/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "social-post", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Hao0321/claude-skill-social-post --skill social-post -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install Hao0321/claude-skill-social-post social-post --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Hao0321/claude-skill-social-post.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/social-post .cursor/skills/social-post && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "social-post" agent skill from https://github.com/Hao0321/claude-skill-social-post/tree/main/social-post into .cursor/skills/social-post/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "social-post", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/Hao0321/claude-skill-social-post.git --path social-post--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add Hao0321/claude-skill-social-post --skill social-post -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install Hao0321/claude-skill-social-post social-post --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Hao0321/claude-skill-social-post.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/social-post .gemini/skills/social-post && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "social-post" agent skill from https://github.com/Hao0321/claude-skill-social-post/tree/main/social-post into .gemini/skills/social-post/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "social-post", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install Hao0321/claude-skill-social-post social-postInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add Hao0321/claude-skill-social-post --skill social-post -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/Hao0321/claude-skill-social-post.git skills-src && mkdir -p .github/skills && cp -r skills-src/social-post .github/skills/social-post && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "social-post" agent skill from https://github.com/Hao0321/claude-skill-social-post/tree/main/social-post into .github/skills/social-post/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "social-post", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Hao0321/claude-skill-social-post --skill social-post -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install Hao0321/claude-skill-social-post social-post --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Hao0321/claude-skill-social-post.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/social-post .opencode/skills/social-post && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "social-post" agent skill from https://github.com/Hao0321/claude-skill-social-post/tree/main/social-post into .opencode/skills/social-post/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "social-post", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
social-post- GitHub stars
- 726
- Token cost
- ~2.9k tokens
- SKILL.md length
- 826 words
- Files
- 231 (incl. scripts, references)
- Skills in repo
- 2
- Repo updated
- First seen
- Licence
- MIT
At a glance
依使用者真實貼文與成效寫 Facebook／Instagram／YouTube／Threads／X 文案，包含 ChatGPT Chat 的「寫文」「Mode C」「用我的格式／口氣」「黑底白字」；本機工作台、規劃、確認後發布、留言回覆及成效學習。使用者說「發文」「文案」「Social Post 介面」「回覆留言」「查流量」「把數據訓練進去」「比較貼文」「優化 pattern」時使用。
- Works in 8 steps: 每組洞察圖建立新 snapshot；不覆蓋舊數字。既有 event… → 每篇新 post 先建立… → complete post… → …
- Tasks that involve Social media posts
- SKILL.md covers Session 啟動, 路由, Source of truth and 流量與演算法單一歸檔中樞, plus 5 more sections
- Calls python and node
What it does
Social Post is an agent skill from Hao0321/claude-skill-social-post. 依使用者真實貼文與成效寫 Facebook／Instagram／YouTube／Threads／X 文案，包含 ChatGPT Chat 的「寫文」「Mode C」「用我的格式／口氣」「黑底白字」；本機工作台、規劃、確認後發布、留言回覆及成效學習。使用者說「發文」「文案」「Social Post 介面」「回覆留言」「查流量」「把數據訓練進去」「比較貼文」「優化 pattern」時使用。
Its SKILL.md is about 2.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 234 other files, including scripts and reference files (for example `.social-post-managed.json`, `agents/openai.yaml` and `comment-capabilities.json`).
It sits in Writing & Content, covering Social media posts. It works with Instagram, OpenAI, YouTube and Model Context Protocol. The repository describes itself as: A Claude Code skill by Hao (駱君昊) that learns your Facebook voice and auto-posts to FB / IG / Threads / X with a 14-day content calendar. Mega-viral validated: 80K reach / 448… The licence is MIT.
When your agent uses it
- Tasks that involve Social media posts
Example prompts
- “/social-post”
Workflow steps
8 steps, taken from the first numbered list in SKILL.md.
- 每組洞察圖建立新 snapshot；不覆蓋舊數字。既有 event 的發布時間、片長或精度要修正時，追加 correction event，不直接改舊列。
- 每篇新 post 先建立 analysis_status、analysis_version、analysis_eligible 與精確 caption 的 caption_sha256；尚未完成語義分析時只能是…
- complete post 必須保存完整文案／版型／長度、換行與段落、標點、關鍵字、named entities、數字語言、opening、hook、mechanism、differentiator、proof、語氣、CTA，以及由 published_at＋timezone…
- caption 截斷、僅有代稱、placeholder 或尚待補證據時標成 excluded_placeholder 或 pending，一律 analysis_eligible:false；缺值用 null／空清單與低 confidence 明示，不補寫不存在的…
- 記 published_at、captured_at、hours_since_publish 與 maturity。IG／FB total 與可取得的拆分同時保存；missing 用 null。
- UI rate 與 derived rate 分開；留存曲線目測只寫 note。縮寫、上限或目測數字用 metric_qualifiers 標成 rounded／upper_bound／visual_estimate，不得冒充 exact。metrics…
- 先 dry-run scripts/log_outcome.py，明確寫入時才加 --write；寫完執行 scripts/social_data.py validate，caption hash、deterministic counts、日期／星期／分鐘或 analysis…
- 新數據不自動改寫作者契約。第三方資料用 social_governance.py record-reference，不得冒充自有 post；圖文與影片分區未知或衝突時不納入學習。新的寫作／F 改善建議用候選審核流程，不直接改 voice 或 F 原檔。
What it can do on your machine
Read from SKILL.md and the folder at commit e0720ed. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
pythonnode
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Social Post loads about 2.9k tokens when it runs, and up to ~40k if it reads all its reference files. Until then it costs about 52 tokens; SKILL.md has 826 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from Hao0321/claude-skill-social-post at commit e0720ed, republished under its MIT licence (© Hao0321). 826 words, ~2,898 tokens.
Download SKILL.mdSave it as
.claude/skills/social-post/SKILL.md (or your agent's skills folder). This skill also uses 230 other files; get the full folder from GitHub.- name
- social-post
- description
- 依使用者真實貼文與成效寫 Facebook／Instagram／YouTube／Threads／X 文案，包含 ChatGPT Chat 的「寫文」「Mode C」「用我的格式／口氣」「黑底白字」；本機工作台、規劃、確認後發布、留言回覆及成效學習。使用者說「發文」「文案」「Social Post 介面」「回覆留言」「查流量」「把數據訓練進去」「比較貼文」「優化 pattern」時使用。
Social Post
把內容生成、實際發布與成效學習分開。依意圖只讀必要資料，不把整個案例庫一次塞進 context。
P5 執行前核對 references/chrome-comment-adapter.md 的當前狀態。CUA Threads 單則已完成真實送出與唯讀對帳；CUA IG 候選已接入但尚無新 runtime 送出實證；FB 送出與三平台批次仍未完成，泛用 production 關閉。target-only 不代表整篇／回覆串已掃完，semantic UI continuity 不代表實體 document epoch 或持久 node identity。單則實證不升級整體 capability，fixture 或候選程式存在也不能代替真實操作證據。
Session 啟動
P0–P4 先讀 references/writing-governance.md，核對 scripts/social_governance.py status。既有 voice、F 公式、規則正文以私人版本契約鎖定；禁止 P1／P4 或新爆款案例直接覆寫。新分析只能建立同平台／媒體 scope 的候選，經使用者審核精確 diff／digest，另行啟用後才變成正式有效版本。當輪要求改一篇草稿不是永久規則更新授權。
P2 必須讀 context 輸出的完整有效 F 與 voice，按固定段落功能、順序、標點與 CTA 寫，交付前核對同 revision 的 check-draft 並完成語義／事實 QA。F06 必須分清 F06a／F06b。結構 PASS 不等於模仿成功；未核對、契約漂移或 F 結構不符不得聲稱正式文案已完成。
P2 預設讀 voice_quick.md；只有 P1 重新學語氣、使用者明確要求深度仿寫，或 quick card 無法裁決時，才完整讀 style_profile.md 或使用者明確指定的 voice Skill。ChatGPT Chat 若沒有本機檔案／shell，依 references/generate_and_publish.md 的降階流程；只讀到公開範本或 GitHub repo，不代表已載入使用者的私人 voice。安全與使用者明示 > voice quick／明確指定的 voice Skill > 公式。
路由
| 觸發 | Mode | 必讀 |
|---|---|---|
| 開啟 Social Post 工作台／介面、看模式導覽 | Workbench | references/workbench.md；只啟動本機 GUI，不暗中執行 AI 或解鎖外部操作 |
| 重新規劃、排內容 | P0 Plan | references/phase0_plan.md＋current_brief.md＋scripts/social_data.py comparables compact context＋目標 formula；X 題材另讀 references/x.md 的演算法邊界 |
| 重新學語氣 | P1 Learn Voice | references/learn_style.md＋style_profile.md |
| 寫一篇、PO、發文 | P2 Draft／Publish | references/generate_and_publish.md＋voice_quick.md＋current_brief.md＋scripts/social_data.py comparables compact context＋單一 formula；X 草稿先讀 references/x.md 的策略段，實際發布前再核對平台 UI |
| 把數據訓練進來、記錄成效 | P3 Log Outcome | references/outcome-workflow.md＋data/*.jsonl |
| 比較貼文／集數、找 pattern | P4 Optimize Patterns | references/outcome-workflow.md＋references/evaluation.md＋相關 rules；X 演算法解釋另讀 references/x.md |
| 掃描、草擬、回覆 FB／IG／Threads 留言 | P5 Comment Ops | references/comment-operations.md＋實際操作時的 references/chrome-comment-adapter.md＋scripts/comment_chrome_actuator.mjs＋references/comment-policy.json＋目標平台 ref＋voice_quick.md |
| 查歷史 Case | Legacy Case | references/case_studies.md 索引，再讀單一 references/cases/case-NN.md |
路由前用一句話告知正在做哪個 Mode。單純診斷不需要 Chrome。
工作流程 P0–P5、寫法 A／B／C、創作公式 F 是三個不同選擇。GUI 只列本機已安裝的公式；不把作者私人公式庫公開。純文字撰稿可直接在當前聊天依 P2 完成，不強迫每次開 GUI。介面目前用任務複製／貼回交接 AI，不能宣稱按鈕已自動呼叫 Codex／Claude。
Source of truth
| 資料 | Canonical source |
|---|---|
| 貼文／caption／發布條件 | data/posts.jsonl |
| 洞察快照 | data/insight_snapshots.jsonl |
| 帳號期間總覽 | data/account_snapshots.jsonl |
| 跨篇假設與 confound | data/experiments.jsonl |
| 事實修正事件 | data/corrections.jsonl；原始 event 不覆寫 |
| 圖文／影片分區 | scripts/social_cohorts.py 的平台 × media family × content type × surface × maturity；不混合排名 |
| 第三方參考／候選與審核 | data/learning_events.jsonl；外部觀測與自有 KPI 嚴格分離 |
| 正式有效寫作契約 | data/author_contract.json＋data/author_contracts/；凍結原始來源、審核後 scoped override，保留前版 |
| 規則正文 | references/rules/RNN.md；references/rules.md 是索引 |
| 規則生命週期／實驗 backlink | references/rules/metadata.json |
| 規則機器索引 | data/rule_registry.json（生成檔） |
| 舊案例全文 | references/cases/ |
| Chrome 可見留言 observation | data/comment_events.jsonl；私人 append-only ledger |
| Chrome 掃描目標／期限 | data/browser_scan_requests.jsonl；私人 append-only ledger，先建立再掃描 |
| 回覆草稿／permit／送出／對帳 | data/reply_events.jsonl；私人 append-only ledger |
| 留言自動化政策 | references/comment-policy.json |
| 產品能力完成度／外部 canary | 私版 .rd/capability-ledger.json 是唯一 canonical ledger；comment-capabilities.json 只是 hash-bound 生成式公開投影，fixture 與 live 分開 |
新成效不得只寫進 Markdown。先寫 JSONL，再視需要更新人類摘要。
流量與演算法單一歸檔中樞
- 不論素材由
video-autopilot、YouTube 專門 Skill、Meta 面板或其他流程產生，所有發布後的流量、推薦來源、搜尋、留存、互動、受眾、追蹤／訂閱與轉化證據，都必須寫回本 Skill 的data/*.jsonl。 - 專門 Skill 可以做平台診斷或產生剪輯／包裝建議；但觀測事實、跨平台比較、實驗證據與規則升降級，以本 Skill 為 canonical source，禁止另建互相漂移的成效記憶。
- 同一內容的 FB／IG／YouTube 指標各自建立有平台 scope 的 snapshot；Meta 合併卡片只能作 cross-platform reference，沒有完整平台洞察時不得冒充該平台 snapshot。
- 演算法時效性知識仍可放目標平台 reference；私人實測是否成立，只能依本 Skill 的同 maturity、多樣本資料升級。
P3 Log Outcome
- 每組洞察圖建立新 snapshot；不覆蓋舊數字。既有 event 的發布時間、片長或精度要修正時，追加 correction event，不直接改舊列。
帳號 7／30／90 天總覽寫
account_snapshots.jsonl，不得綁到單篇貼文。 - 每篇新 post 先建立
analysis_status、analysis_version、analysis_eligible與精確 caption 的caption_sha256；尚未完成語義分析時只能是pending＋analysis_eligible:false，不得先進 pattern learning。 completepost 必須保存完整文案／版型／長度、換行與段落、標點、關鍵字、named entities、數字語言、opening、hook、mechanism、differentiator、proof、語氣、CTA，以及由published_at＋timezone 驗證出的日期、星期、HH:mm與 daypart。黑底白字另存背景、字色、字重、對齊與觀測行數；影片另存直式版型，不從縮圖猜未見內容。- caption 截斷、僅有代稱、placeholder 或尚待補證據時標成
excluded_placeholder或pending，一律analysis_eligible:false；缺值用null／空清單與低 confidence 明示，不補寫不存在的 hook、關鍵字、語氣或 CTA。 - 記 published_at、captured_at、hours_since_publish 與 maturity。IG／FB total 與可取得的拆分同時保存；missing 用
null。 - UI rate 與 derived rate 分開；留存曲線目測只寫 note。縮寫、上限或目測數字用
metric_qualifiers標成rounded／upper_bound／visual_estimate，不得冒充 exact。metrics內的值直接用欄位名；其他巢狀值使用 dotted path（例如benchmarks_reported.views_more_than_usual_approx）。 - 先 dry-run
scripts/log_outcome.py，明確寫入時才加--write；寫完執行scripts/social_data.py validate，caption hash、deterministic counts、日期／星期／分鐘或 analysis eligibility 任一不一致都不得通過。 - 新數據不自動改寫作者契約。第三方資料用
social_governance.py record-reference，不得冒充自有 post；圖文與影片分區未知或衝突時不納入學習。新的寫作／F 改善建議用候選審核流程，不直接改 voice 或 F 原檔。
P4 Optimize Patterns
先跑 feature matrix；series summary 只作 KPI 總覽。Matrix 只納入 analysis_status:complete、analysis_eligible:true、caption hash 與 analysis version 均有效的 post，並在選 latest 前排除 awaiting_data／aggregation_eligible:false，再以相同平台 scope 與 maturity 比較。每列必須同時帶出文案長度、段落／分隔、完整 Unicode 標點 profile、關鍵字、語氣、CTA、黑底白字／影片版型、日期、星期、分鐘與成效；可用 --captured-before 重建當時可見的歷史比較，避免後來快照污染判斷。
跨平台同步內容若有可驗證的各平台收據，使用 platform_publications 分別記錄平台發布時間、IANA timezone、sync mode、caption digest 與 media digest；只有共同時間、沒有逐平台憑證時保留單一 published_at 並明示信心，不虛構每個平台的分鐘。不同 UI 表面出現 104.2／104.3 或 211／215 這類差異時，保留全部 surface_observations、表面與證據，canonical 值只作查詢基準，不覆蓋原始差異。
故事、首幀、集數、caption、CTA、版型或時段同時改變時，列為 confound，不稱乾淨 A/B。發布時間只可作候選變因：同一精確分鐘若同時出現高低差異很大的結果，不得把該分鐘寫成爆款原因或固定黃金時段；必須由相近題材、包裝、平台與 maturity 的獨立樣本另行驗證。
證據狀態只用：hypothesis → emerging → validated → deprecated。同一系列三集是 n=3 posts，但不是三個獨立樣本。實驗可用同一 experiment_id 追加 revision；不可覆寫歷史。候選規則必須同時在 experiment rule_ids 與 rule metadata experiment_ids 建 backlink。
Show full SKILL.md (379 more words)Show less
實際發佈與留言安全閘
只有 P2 的實際發布與 P5 的實際掃描／回覆需要已登入的 Chrome UI 工具；P5 依 runtime reference 使用當前文件化 CUA，舊 chrome:control-chrome 僅限仍提供對應 SDK 的環境。草稿、規劃、分析、資料回填與 P5 ledger 操作不需要瀏覽器。
- 發佈前必須在當前對話取得明確「確認」。
- 使用者若在當前 session 明示「你自己操作不用問」，私人版可免逐次確認；不跨 session。
- 不幫登入、不改帳號／隱私、不刪文、不自動按讚／follow／大量留言。
- P5 完整掃描前先以當前 session 建立有期限的 browser scan request，Chrome receipt 只能回綁既定帳號／貼文 scope，完成掃描即追加可區分零留言的 completion event。production scan 模組不含 fixture factory，只接受 source-controlled、trusted-host-resolved、版本化、deep-frozen 且 process-branded 的 FB／IG／Threads plan；目前泛用完整掃描 registry 全部 unavailable，指定留言 target-only 回填是獨立範圍。完整展開必須由 versioned exhaustion contract 證明 cursor traversal、monotonic discovered count、explicit terminal 與 terminal coverage；兩次空 viewport read、raw selector、fake tab、caller resolver 或
threadExpansionComplete宣告都無效。P5 預設batch_confirm；bounded_auto只在當前 session 明示平台、帳號、貼文與本輪範圍後，以有期限、指定 scope、有限次數的 ledger grant 啟用。每則回覆都要一次性 permit；重算 reply hash，綁定 action／fresh locator／正確父留言與零 exact-own baseline。泛用送出契約在碰 trigger、消耗 durable claim、finish 與 recovery reinspection 前，另以 version 1 exhaustion 證明 target-scoped cursor／count／traversal／terminal stable coverage；0reply 或0expander 不構成 absence。later page／lazy expander、無 terminal、virtualization 或 stable-node replacement 不能當泛用完整證據，claim 前發現則零 claim／submit，recovery 不得推導 not-sent。Threads 單則候選另用原生 explicit-zero marker、兩次穩定讀取與來源點選的空 modal；只提供 semantic selection，不取得泛用 complete／absence 或持久節點 authority。之後才可經 durable atomic claim、process-wide 單次送出及原生結果驗證。finish／reconcile 另須由 shell:false branded bridge 持有 versioned 一次性 capability；ledger 只存 nonce hash、scope binding 與 receipt digest。裸WRITE_OK、raw／fixture receipt、偽 nonce、重播 capability，或冒充完整掃描的部分回覆串一律拒絕；分批執行不得重置 grant 上限。 - 正式版
comment-policy.json的live_browser_actuation_enabled預設為false；來源綁定的只讀回填另受live_browser_scan_enabled控制。泛用送出仍關閉；單則候選只可使用當前 session 明確授權、綁定已核准 action／source digest、最長 300 秒且限一則的 canonical canary lease，經來源持有的executeCanaryReply執行。CUA 接入 Threads 與 IG 的單則 preparation，但實證狀態須分開核對。lease 不開啟 production、不升級 capability、不允許 caller 偽造 preparation；IG 的@author原生帶入與 Threads 的真正空 modal 分別如實記錄。成功只認正確父留言下的新 exact-own 回覆及有效帳本結算，未知立即停止且不得重送。對帳只展開／讀取，不得走送出前的選取或填字流程。 - 送出結果不明時標記
needs_reconcile並停止整批；未重新讀取畫面前不得重送。零 API Chrome 模式不宣稱 24/7 背景監聽。 send_started／needs_reconcile後若 Node／Chrome 重啟或 receipt capability 過期，只能由 fused actuator 以 fresh session 進入 reconcile recovery；bearer 留在私有 Node 閉包，保留原 attempt、不得重發 submit claim、不得退回approved。- P5 不處理私訊、媒體／GIF 回覆或全帳號歷史爬取；大量 keyword 索取改用單一公開作者留言提供自助入口。
- 預設跨平台重新包裝；但使用者明示「同步發布／一稿多發」時，正文共用一份，只有平台必填欄位沿用正文內容（例如 YouTube 標題取第一句），不再額外維護多套文案。
- FB／Threads 正文不放外部連結；依 R25 使用留言或平台允許的位置。
- 沒有 IG 圖／影片就停，讓使用者選擇提供素材、跳過 IG 或改 Threads。
平台規則
平台規格會變。Hashtag、字數、發佈 UI、演算法等時效規則只在目標平台 reference 維護，標示 last verified；跨 skill 衝突時先查權威來源，不同時保留兩個硬數字。X 的 For You 機制、來源版本與可觀測邊界見 references/x.md；不得把原始碼的預測權重當成實際互動兌換率或保證觸及的寫作公式。
UI 改版只更新已授權平台的 reader／selection／result 模組及相應測試，不重寫或放寬共用 ledger、permit 與單次 claim 契約。權限拒絕、CAPTCHA、checkpoint 或平台限制必須暫停，不以換 URL、runtime、帳號或自動重試繞過，也不保證不中斷運作。
維護
- 每次 outcome 更新後跑 data validate、rule registry build、cleanup drift audit。
- 每次 outcome／analysis 更新後另跑
python scripts/social_data.py coverage；只有coverage_complete=true，且每個適用的平台 outcome 子樹都通過獨立 exact-compare 維度，才可宣稱每篇 eligible post 的文案、長度、版型、關鍵字、語氣、完整標點、日期／星期／分鐘與全部 outcomes 已進入可供 P0／P2 使用的矩陣。固定欄位以外的新平台 analytics 自動進extended_analytics，不可因 schema 尚未命名而遺漏。 - 修改任一
rules/RNN.md後跑split_rule_archive.py --refresh-manifest，再 build rule registry。 case_studies.md只保留索引；新增 Case 寫獨立檔或直接以 structured outcome 取代。- 公開 export 使用 allowlist：只公開通用引擎、匿名範例與去快照操作卡；不得公開完整
style_profile.md、content_plan.md、drafts/、任何 outcome／correction JSONL、私人 rules／formulas／cases 或.rd/。同步工具必須先跑 privacy preflight，BLOCK 時不得寫入鏡像。 - 維護公開版時，先同步、再驗證公開鏡像；只有當輪使用者明確授權 release 時，才交由 R&D external-change gate 推送。不得把一次授權永久化。
- 能力狀態只改
.rd/capability-ledger.json，再用python scripts/comment_capability_gate.py --write-projection重建公開留言投影；gate 必須拒絕漏項、額外 obligation、狀態漂移、來源 hash 漂移與手改 projection。 - 私公版路徑只從
audit.config.json讀；未設定就回報 NOT_CHECKED，不猜 sibling repo。 - 修改 Chrome comment JavaScript 後，必須先跑
node scripts/comment_js_architecture_gate.mjs --self-test，再以--output .rd/receipts/js-architecture-gate.json保存 actual closed-graph receipt；最後跑comment_self_test.py。此產品原生 gate 只補 JS graph 證據，不得把 Cleanup 的跨語言NOT_CHECKED改寫成 provider PASS。
$env:PYTHONUTF8='1'
python scripts/social_data.py validate
python scripts/social_data.py coverage
python scripts/social_data.py summary --series <series-id>
python scripts/social_data.py matrix --series <series-id> --platform <platform> --maturity <maturity>
python scripts/build_rule_registry.py --write
python scripts/comment_assistant.py validate
node scripts/comment_js_architecture_gate.mjs --self-test
node scripts/comment_js_architecture_gate.mjs --output .rd/receipts/js-architecture-gate.json
python scripts/comment_self_test.py
python ../code-cleanup-helper/scripts/audit.py . --mode all© Hao0321, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 230 other files (scripts, references) in social-post of Hao0321/claude-skill-social-post.
- SKILL.md
- .social-post-managed.json
- agents/openai.yaml
- comment-capabilities.json
- content_plan.example.md
- current_brief.md
- data/rule_registry.json
- references/case_studies.md
- references/cases/case-01.md
- references/cases/manifest.json
- references/chrome-comment-adapter.md
- references/chrome-comment-runtime.md
- references/chrome-comment-validation.md
- references/comment-operations.md
- references/comment-policy.json
- references/correction-bundle.example.json
- references/evaluation.md
- … and 214 more
Open the folder on GitHubat commit e0720ed
Compare with similar skills
Social Post next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Social Post this skillHao0321/claude-skill-social-post | 726 | — | ~2.9k | Automated safety check: Pass | MIT | |
| Outlier Post FinderScrapeCreators/social-media-research-skills | 3.3k | — | ~1.5k | Automated safety check: Notes | MIT | |
| Content Repurposer Smsblacktwist/social-media-skills | 557 | — | ~3.7k | Automated safety check: Pass | MIT | |
| Social Media Managementmanojbajaj95/claude-gtm-plugin | 104 | 1 repos | ~3.9k | Automated safety check: Pass | MIT | |
| Hook Writer Smsblacktwist/social-media-skills | 557 | — | ~3.7k | Automated safety check: Pass | MIT | |
| Thread Writer Smsblacktwist/social-media-skills | 557 | — | ~5k | Automated safety check: Pass | MIT |
Similar skills
Content Repurposer Sms
blacktwist/social-media-skills
When the user wants to turn one piece of content into multiple formats or adapt content across text-first and visual-first platforms (LinkedIn, Twitter/X, Threads, Bluesky, Facebook, Instagram…
Social Media Management
manojbajaj95/claude-gtm-plugin
Comprehensive social media management for all platforms (LinkedIn, Twitter/X, Instagram, TikTok, Facebook, Pinterest, YouTube).
Hook Writer Sms
blacktwist/social-media-skills
When the user wants help writing opening lines, hooks, first sentences, video hooks, thumbnails titles, or pin titles that grab attention.
Thread Writer Sms
blacktwist/social-media-skills
When the user wants to write a multi-part thread or content series for Twitter/X, LinkedIn, Threads, Instagram (Reel/carousel/Story series), TikTok (multi-part videos), YouTube (video series…
AI Social Media Content
NeverSight/learn-skills.dev
Create AI-powered social media content for TikTok, Instagram, YouTube, Twitter/X.
More from Hao0321/claude-skill-social-post
Socialpost
Hao0321/claude-skill-social-post
在 ChatGPT 聊天中規劃、撰寫與修改 Facebook、Instagram、Threads、YouTube、X 貼文；處理 Mode C、黑底白字、作者語氣、版型與成效證據。使用者說「發文」「寫文」「文案」「用我的口氣／格式」「Mode C」「黑底白字」「分析貼文」時啟用。
Related
Categories
Questions about Social Post
What does Social Post do?
依使用者真實貼文與成效寫 Facebook／Instagram／YouTube／Threads／X 文案，包含 ChatGPT Chat 的「寫文」「Mode C」「用我的格式／口氣」「黑底白字」；本機工作台、規劃、確認後發布、留言回覆及成效學習。使用者說「發文」「文案」「Social Post 介面」「回覆留言」「查流量」「把數據訓練進去」「比較貼文」「優化 pattern」時使用。. Social Post is an agent skill from Hao0321/claude-skill-social-post.
When should I use Social Post?
Social Post fits situations like: tasks that involve Social media posts.
How do I install Social Post in Claude Code?
Run `npx skills add Hao0321/claude-skill-social-post --skill social-post -a claude-code`. Or copy the skill folder (social-post in Hao0321/claude-skill-social-post) into .claude/skills/social-post in your project. Claude Code loads it when a task matches its description.
How do I install Social Post in Codex?
Run `npx skills add Hao0321/claude-skill-social-post --skill social-post -a codex`. Or copy the skill folder (social-post in Hao0321/claude-skill-social-post) into .agents/skills/social-post in your project. Codex loads it when a task matches its description.
Can I use Social Post in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Hao0321/claude-skill-social-post --skill social-post -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/social-post, .gemini/skills/social-post, .github/skills/social-post and .opencode/skills/social-post in your project.
What does Social Post need to run?
Going by SKILL.md and its folder, Social Post needs the command-line tools its instructions call (python and node).
Does Social Post access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Social Post safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Social Post use?
Social Post is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Social Post use?
About 2.9k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 37k tokens, read only when the agent opens those files.
What are the alternatives to Social Post?
Skills that share tags, products or a category with Social Post: Outlier Post Finder (ScrapeCreators/social-media-research-skills, 3.3k stars), Content Repurposer Sms (blacktwist/social-media-skills, 557 stars), Social Media Management (manojbajaj95/claude-gtm-plugin, 104 stars) and Hook Writer Sms (blacktwist/social-media-skills, 557 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Social Post?
Hao0321 (a GitHub user) maintains it in Hao0321/claude-skill-social-post, which has 726 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on October 3, 2026.
Source: Hao0321/claude-skill-social-post on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.