本文へ移動
cccskills
無料GitHub で公開

taisly-social-media-posting

Free AI-first short-form video publishing to TikTok, Instagram Reels, YouTube Shorts, X, and Facebook from AI agents through Taisly.

インストール方法を見る

含まれるファイル(5)

  • SKILL.md6.0 KB
  • .codex-plugin/plugin.json1.6 KB
  • LICENSE1.0 KB
  • package.json1.7 KB
  • README.md13.1 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Taisly Social Media Posting Skill

Use this skill when a user asks an AI agent to publish, schedule, or repost short-form video content through Taisly. The package is free to install and designed for AI-first posting workflows that can start without building platform integrations from scratch.

Rules

  • Always discover platforms before posting.
  • Ask for explicit confirmation before publishing to live social accounts unless the user already gave exact destination accounts, video file, caption, and schedule.
  • If MCP tools are available, prefer them over shell commands; otherwise use the JSON-first CLI.
  • For MCP publishing, call taisly_posts_create only with confirmed: true after explicit user approval.
  • Never invent platform IDs.
  • Never expose or print TAISLY_API_KEY.
  • If Taisly returns a current-plan or limit error, explain agent.message when present and provide the returned agent.paymentLinks when available. Do not attempt payment without explicit user confirmation.
  • Prefer scheduled posts when the user gives a future time.
  • Save the returned historyId so status can be checked later.
  • Supported local video extensions are .mp4, .mov, .avi, .mkv, .webm, .flv, .mpeg, and .mpg.

Environment

Prefer the browser setup flow when the user has not configured Taisly yet:

taisly setup --agent <agent-slug>
# Ask the user to open the returned loginUrl and finish authentication.
taisly checkin --agent <agent-slug>

Use a short slug for the current agent, for example claude-code, codex, cursor, windsurf, openclaw, hermes-agent, cline, or aider.

The checkin command stores a local Taisly agent credential for later CLI and MCP calls.

Manual API key fallback:

export TAISLY_API_KEY="taisly_..."

Remote MCP endpoint:

https://app.taisly.com/mcp

Remote MCP can publish only from public videoUrl values. Use the local MCP server or CLI when the video is a local file path available to the agent.

Commands

taisly auth:status
taisly setup --agent <agent-slug>
taisly checkin --agent <agent-slug>
taisly platforms:list
taisly integrations:list
taisly platforms:schema --platform TikTok
taisly posts:validate --video ./video.mp4 --platforms platform_id_1,platform_id_2 --description "Caption"
taisly posts:create --video ./video.mp4 --platforms platform_id_1,platform_id_2 --description "Caption" --scheduled "2026-06-14T09:00:00+07:00"
taisly posts:create --json ./campaign.json
taisly posts:list --page 1
taisly posts:status --id <historyId>
taisly reposts:create --from <platform_id> --to <platform_id_1,platform_id_2>
taisly mcp

All commands return JSON.

MCP Tools

When the MCP server is connected, use these tools instead of shell commands:

  • taisly_agent_setup_start
  • taisly_agent_checkin
  • taisly_auth_status
  • taisly_platforms_list
  • taisly_platform_schema
  • taisly_platform_connect_start
  • taisly_platform_connect_check
  • taisly_posts_validate
  • taisly_posts_create
  • taisly_posts_status
  • taisly_posts_list
  • taisly_reposts_list
  • taisly_reposts_create

Posting Workflow

  1. If Taisly is not connected yet, run taisly_agent_setup_start or taisly setup --agent <agent-slug> and give the user the returned login URL.
  2. Wait for the user to finish browser login, then run taisly_agent_checkin or taisly checkin --agent <agent-slug>.
  3. Run taisly auth:status.
  4. Run taisly platforms:list.
  5. If a requested platform is missing and a connect tool is available, run taisly platforms:connect:start --platform <name>, give the user the returned connectUrl, wait for browser approval, then run taisly platforms:connect:check --platform <name>. Supported connect slugs are instagram, tiktok, youtube, x, and facebook.
  6. Match the user's requested platforms to connected platform IDs.
  7. Run taisly platforms:schema --platform <name> for constraints.
  8. Run taisly posts:validate.
  9. Confirm destination accounts and caption with the user.
  10. Run taisly posts:create.
  11. Report the returned historyId, scheduled date, and per-platform initial statuses.

For MCP, follow the same sequence with taisly_auth_status, taisly_platforms_list, taisly_platform_connect_start, taisly_platform_connect_check, taisly_platform_schema, taisly_posts_validate, taisly_posts_create, and taisly_posts_status.

Error Handling

  • TAISLY_API_KEY_MISSING: ask the user to create or provide an API key.
  • SETUP_SESSION_MISSING: run taisly setup --agent <agent-slug>, ask the user to open the returned login URL, then run taisly checkin --agent <agent-slug>.
  • PLATFORMS_REQUIRED: ask which connected accounts should receive the post.
  • VIDEO_REQUIRED: ask for a local video path.
  • LIMIT: tell the user the current plan limit has been reached. Use agent.message and agent.paymentLinks when present, ask them to upgrade in Taisly before retrying, and do not open or use payment links without explicit confirmation.
  • POST_NOT_FOUND_IN_RECENT_HISTORY: tell the user the post was created but a dedicated status endpoint is not available yet; check Taisly History.

What Not To Do

  • Do not post to every connected account unless the user explicitly asks.
  • Do not retry posts:create blindly after a timeout; ask the user or check History first.
  • Do not use unsupported file types.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Use AtomLane to compile and execute safe atomic parallel plans on macOS and native Windows Preview for worthwhile independent argv tasks, dependency DAGs, supported platform entrypoints, or Apple-silicon operators. Use at task start or an execution boundary when structured local work may contain two or more worthwhile units; skip plain answers, one quick command, and work whose effects cannot be safely bounded.

日本語の概要は準備中です。原文の説明を表示しています。

hashgraph-online/awesome-codex-plugins1,2932026年10月13日 更新

add

無料

Register a deferred decision in the debt registry. Trigger by judgment, not a marker scan, whenever a future reader would ask "why this way?": an unmade decision, stub, loosened type, bypassed check, swallowed error, a default picked "for now", or a TODO/FIXME/HACK/XXX marker. Trigger immediately whenever you defer work, or when the user invokes $add. Over-register freely; the developer drops with "drop A", "drop A,C", or "drop all".

日本語の概要は準備中です。原文の説明を表示しています。

hashgraph-online/awesome-codex-plugins1,2932026年10月13日 更新

ADK 框架适配层。为 LangChain / EINO / AutoGen / AgentScope / CrewAI 提供框架特定的 代码模板、惯用模式、API 映射和项目结构,供 agent-dev-workshop Phase 5 代码生成使用。 每个框架 reference 文件标注 verified_date 用于版本锁定。

日本語の概要は準備中です。原文の説明を表示しています。

hashgraph-online/awesome-codex-plugins1,2932026年10月13日 更新

中文调试修复技能。用于报错、测试失败、页面异常、功能不符合预期、需要定位根因并做最小修复时。触发语包括"进入调试模式""帮我修问题""报错了""测试失败""页面坏了""找根因"。

日本語の概要は準備中です。原文の説明を表示しています。

hashgraph-online/awesome-codex-plugins1,2932026年10月13日 更新

交互式 AI Agent 开发工作坊:通过 6 阶段深度协作对话,引导用户完成 Agent 需求分析、架构设计、 工具定义、Prompt 与编排设计、代码生成、验证迭代,产出可直接运行的 Agent 项目。 框架无关设计优先,支持 LangChain / EINO / AutoGen / AgentScope / CrewAI 等 ADK 框架。

日本語の概要は準備中です。原文の説明を表示しています。

hashgraph-online/awesome-codex-plugins1,2932026年10月13日 更新

中文漂移审计技能。用于项目或学习过程变乱、上下文漂移、任务分叉、多个方案冲突、命名不一致、Codex 可能顺手改多了时。触发语包括"漂移检查""感觉跑偏了""项目变乱了""检查是否失控""分叉太多""上下文漂移"。

日本語の概要は準備中です。原文の説明を表示しています。

hashgraph-online/awesome-codex-plugins1,2932026年10月13日 更新

hashgraph-online のスキルをすべて見る

このスキルの問題を報告する