Token导航 LogoToken导航

Agent Skill · 研究检索

posta(posta 图像)

posta 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

posta 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 15,674。

总安装 7,680周安装 320GitHub Stars 公开资料未说明下载 2,560核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.2.2

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:posta(posta 图像)
来源仓库:https://github.com/stgime/posta
安装命令:
openclaw skills install posta
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install posta

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于从终端向多个社交平台发布内容,支持图文与视频格式。适合在 OpenClaw 中统一管理 Instagram、TikTok、X 等渠道的发布流程。通过 clawhub 安装后,可使用 AI 生成图像并自动排版发布。需分别授权各平台 API 访问权限,注意内容格式与尺寸适配问题。

建议提前预览发布效果,避免因网络延迟或缓存导致显示异常。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install posta;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

posta 怎么安装?

在终端执行 openclaw skills install posta。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

posta 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

posta 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

posta 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

posta 用的人多吗?

公开数据是累计安装 7,680 次,近一周安装 320 次,下载 2,560 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

posta 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
posta

posta 图像 · 本页

7,680—接触敏感数据MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
posta
description
Post to Instagram, TikTok, LinkedIn, YouTube, X/Twitter, Facebook, Pinterest, Threads and Bluesky from your terminal. Create posts with AI-generated images and captions, upload media, schedule or publish instantly, view analytics, and manage all your social accounts — without leaving your editor. Use this skill when the user wants to create social media content, generate images/videos/text with AI, upload media, create posts, schedule or publish posts, view analytics, compare post performance, or manage social accounts through Posta.
license
MIT
homepage
https://github.com/STGime/posta-skill
metadata
author
Posta
version
1.2.2
tags
clawdbot
requires
env
bins
primaryEnv
POSTA_API_TOKEN

Posta — Social Media Content & Scheduling

Posta is a social media management platform that lets you create, schedule, and publish posts across Instagram, TikTok, Facebook, X/Twitter, LinkedIn, YouTube, Pinterest, Threads, and Bluesky.

This skill enables you to interact with the Posta API to manage social media content end-to-end: authenticate, list accounts, upload media, create/schedule/publish posts, generate AI content, and view analytics.

Setup

Authentication (one of the following)

  • POSTA_API_TOKEN — Recommended. Personal API token (starts with posta_). Long-lived, revocable, no password exposure.
  • POSTA_EMAIL + POSTA_PASSWORD — Legacy login. The skill logs in and caches a JWT automatically.

If POSTA_API_TOKEN is set, email/password are not needed and the login flow is skipped entirely.

Optional Environment Variables

  • POSTA_BASE_URL — API base URL (default: https://api.getposta.app/v1)
  • FIREWORKS_API_KEY — Fireworks.ai API key (for image generation). Keys start with fw_. Get one at https://fireworks.ai/account/api-keys. The skill auto-discovers this from env vars, ~/.posta/credentials, or .env files.
  • GEMINI_API_KEY — Google Gemini API key (for caption/text generation)
  • OPENAI_API_KEY — OpenAI API key (alternative text generation)

Credentials Auto-Discovery

The skill searches a fixed list of dedicated config files for POSTA_API_TOKEN (or legacy POSTA_EMAIL/POSTA_PASSWORD). Only exact variable names are matched — no other file content is read. Shell profiles (~/.zshrc, ~/.bashrc) are never accessed. Search order:

  1. Already-set environment variables (no file access)
  2. ~/.posta/credentials — dedicated Posta config file (preferred)
  3. .env, .env.local, .env.production in the working directory

If POSTA_API_TOKEN is found, the skill uses it immediately and skips email/password lookup. See SECURITY.md in the repo root for full details.

Helper Script

Source the bash helper for all API interactions:

source "${POSTA_SKILL_ROOT:-${OPENCLAW_SKILL_ROOT:-${CLAUDE_PLUGIN_ROOT:-}}}/skills/posta/scripts/posta-api.sh"

This provides:

  • Auth & Core: posta_login, posta_get_token, posta_api, posta_discover_credentials
  • Media: posta_detect_mime, posta_upload_media, posta_upload_from_url, posta_list_media, posta_get_media, posta_delete_media, posta_generate_carousel_pdf
  • Posts: posta_list_posts, posta_create_post, posta_create_post_from_file, posta_get_post, posta_update_post, posta_delete_post, posta_schedule_post, posta_publish_post, posta_cancel_post, posta_get_calendar
  • Platform Discovery: posta_list_platforms, posta_get_platform_specs, posta_get_aspect_ratios, posta_get_platform, posta_get_pinterest_boards
  • Analytics: posta_get_analytics_overview, posta_get_analytics_capabilities, posta_get_analytics_posts, posta_get_post_analytics, posta_get_analytics_trends, posta_get_best_times, posta_get_content_types, posta_get_hashtag_analytics, posta_compare_posts, posta_get_benchmarks, posta_export_analytics_csv, posta_export_analytics_pdf, posta_refresh_post_analytics, posta_refresh_all_analytics
  • User: posta_get_plan, posta_get_profile, posta_update_profile
  • Fireworks: fireworks_validate_key

Reference Docs


Core Workflows

1. Authenticate

Authentication is automatic. If POSTA_API_TOKEN is set, the skill uses it directly — no login step needed. Otherwise it falls back to email/password login with JWT caching. If a request returns 401:

  • API token: reports the token is invalid/revoked (no retry)
  • JWT: re-authenticates and retries once
source "${POSTA_SKILL_ROOT:-${OPENCLAW_SKILL_ROOT:-${CLAUDE_PLUGIN_ROOT:-}}}/skills/posta/scripts/posta-api.sh"
# Token is fetched/cached automatically on first API call

To verify credentials are working:

posta_api GET "/auth/me"

2. List Connected Social Accounts

ACCOUNTS=$(posta_list_accounts)
# Returns a plain array (wrapper is auto-unwrapped)
echo "$ACCOUNTS" | jq -r '.[] | "\(.platform)\	\(.username)\	\(.isActive)"'

Display as a table showing: Platform, Username, Active status, Last used.

Note: Account IDs from posta_list_accounts are integers (e.g. 35). Wrap them in quotes when passing to socialAccountIds: "socialAccountIds": ["35"]

3. Upload Media

The upload flow has 3 steps: create signed URL → PUT binary → confirm upload. MIME type is auto-detected from the file — no need to specify it manually.

From a local file (auto-detect MIME):

MEDIA_ID=$(posta_upload_media "/path/to/file.jpg")

From a local file (explicit MIME):

MEDIA_ID=$(posta_upload_media "/path/to/file.jpg" "image/jpeg")

From a URL (auto-detect from extension):

MEDIA_ID=$(posta_upload_from_url "https://example.com/image.png")

Detect MIME type separately:

MIME=$(posta_detect_mime "/path/to/file.mp4")
# Returns: video/mp4

Supported formats:

  • Images: image/jpeg, image/png, image/webp, image/gif (max 20MB)
  • Videos: video/mp4, video/quicktime, video/webm (max 500MB)

After upload, the media enters processing status. For images this is fast (thumbnails/variants). For videos it takes longer. Check status with:

posta_get_media "$MEDIA_ID"

List media library:

ALL_MEDIA=$(posta_list_media)
IMAGES_ONLY=$(posta_list_media "image")
COMPLETED=$(posta_list_media "" "completed" 50)

Delete media:

posta_delete_media "$MEDIA_ID"

Generate carousel PDF from images:

RESULT=$(posta_generate_carousel_pdf '["media-id-1", "media-id-2", "media-id-3"]' "My Carousel Title")

4. Create, Schedule & Publish Posts

Create a draft post:

POST=$(posta_create_post '{
  "caption": "Your caption here",
  "hashtags": ["tag1", "tag2"],
  "mediaIds": ["media-uuid"],
  "socialAccountIds": ["35", "42"],
  "isDraft": true
}')
POST_ID=$(echo "$POST" | jq -r '.id')

Create a post with multiline caption (from file):

cat > /tmp/caption.txt << 'EOF'
Line one of the caption.

Line two with details.

Call to action here.
EOF
POST=$(posta_create_post_from_file /tmp/caption.txt '["media-uuid"]' '["35", "42"]' true '["tag1", "tag2"]')
POST_ID=$(echo "$POST" | jq -r '.id')

Schedule for a specific time:

posta_schedule_post "$POST_ID" "2026-03-15T09:00:00Z"

Reschedule an already-scheduled post: The API only allows scheduling posts in draft status. To reschedule, cancel first, then schedule again:

posta_cancel_post "$POST_ID"
posta_schedule_post "$POST_ID" "2026-03-16T09:00:00Z"

Publish immediately:

posta_publish_post "$POST_ID"

Platform-specific configuration (optional):

{
  "platformConfigurations": {
    "tiktok": {
      "privacyLevel": "PUBLIC_TO_EVERYONE",
      "allowComment": true,
      "allowDuet": false,
      "allowStitch": false
    },
    "pinterest": {
      "boardId": "board-id",
      "link": "https://your-link.com",
      "altText": "Image description"
    }
  }
}

Note: Either caption or at least one mediaIds entry is required. Text-only posts work for X/Twitter.

5. Generate AI Content

Generate an image with Fireworks SDXL:

curl -s -X POST \
  "https://api.fireworks.ai/inference/v1/image_generation/accounts/fireworks/models/stable-diffusion-xl-1024-v1-0" \
  -H "Authorization: Bearer ${FIREWORKS_API_KEY}" \
  -H "Content-Type: application/json" \
  -H "Accept: image/png" \
  -d '{
    "prompt": "your descriptive prompt, photorealistic, natural colors, high quality, detailed",
    "negative_prompt": "text, watermark, blurry, low quality, distorted",
    "width": 1024, "height": 1024, "steps": 30, "guidance_scale": 7.5
  }' --output /tmp/generated.png

MEDIA_ID=$(posta_upload_media /tmp/generated.png "image/png")

Generate a caption with Gemini:

CAPTION=$(curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key=${GEMINI_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{"parts": [{"text": "Write an engaging Instagram caption about [topic]. Max 150 words."}]}],
    "generationConfig": {"temperature": 0.8, "maxOutputTokens": 300}
  }' | jq -r '.candidates[0].content.parts[0].text')

See content-generation.md for full patterns including OpenAI and hashtag generation.

6. View Analytics

Overview stats:

OVERVIEW=$(posta_get_analytics_overview "30d")
echo "$OVERVIEW" | jq '{totalPosts, totalImpressions, totalEngagements, avgEngagementRate}'

Best posting times:

BEST_TIMES=$(posta_get_best_times)

Top performing posts:

TOP=$(posta_api GET "/analytics/posts?limit=10&sortBy=engagements&sortOrder=desc")

Trends over time:

TRENDS=$(posta_api GET "/analytics/trends?period=30d&metric=engagements")

Check plan and usage:

PLAN=$(posta_get_plan)
echo "$PLAN" | jq '{plan, usage, limits}'

7. Platform Discovery

Query platform capabilities, character limits, media requirements, and supported features before creating posts.

List all supported platforms:

posta_list_platforms

Get full specs (char limits, media requirements, features):

SPECS=$(posta_get_platform_specs)

Get specs for a specific platform:

posta_get_platform "instagram"

Get aspect ratio reference:

posta_get_aspect_ratios

Get Pinterest boards for a connected account:

BOARDS=$(posta_get_pinterest_boards "$ACCOUNT_ID")

Use platform discovery to validate content before posting — check character limits, required media dimensions, and supported post types.

8. Calendar View

View scheduled and posted content on a calendar:

CALENDAR=$(posta_get_calendar "2026-03-01" "2026-03-31")
echo "$CALENDAR" | jq '.items[] | {id, caption: .caption[:50], status, scheduledAt}'

9. Extended Analytics

Analytics capabilities (what your plan supports):

posta_get_analytics_capabilities

Top performing posts (sorted, paginated):

TOP=$(posta_get_analytics_posts 10 0 "engagements" "desc")

Single post analytics:

posta_get_post_analytics "$POST_ID"

Trends over time with custom period:

TRENDS=$(posta_get_analytics_trends "90d" "engagement_rate")

Content type performance breakdown:

posta_get_content_types

Hashtag performance (pro plan):

posta_get_hashtag_analytics

Compare posts side by side (2-4 posts, pro plan):

posta_compare_posts "post-id-1,post-id-2,post-id-3"

Engagement benchmarks (pro plan):

posta_get_benchmarks

Export analytics:

posta_export_analytics_csv "30d"
posta_export_analytics_pdf "90d"

Refresh analytics:

posta_refresh_post_analytics "$POST_RESULT_ID"
posta_refresh_all_analytics  # Rate limited: 1 per hour

Guidelines

  1. Always show a preview before publishing. Display the caption, target platforms, media description, and scheduled time. Ask for confirmation before calling publish or schedule.
  1. Suggest optimal posting times. When the user wants to schedule, fetch best-times analytics and recommend the highest-engagement time slot.
  1. Ask before spending API credits. Image generation (Fireworks) and text generation (Gemini/OpenAI) cost money. Confirm with the user before making generation API calls.
  1. Handle errors gracefully. If an API call fails, show the error message and suggest next steps (check credentials, verify account connection, check plan limits).
  1. Respect plan limits. Check the user's plan with posta_get_plan before attempting operations that may exceed limits (posts, accounts, storage).
  1. Use appropriate aspect ratios. Match the content format to the target platform — portrait for TikTok/Reels, square for Instagram feed, landscape for LinkedIn/X.
  1. Create posts as drafts first. Always set isDraft: true when creating posts, then schedule or publish after user confirmation.
  1. Combine media types strategically. For maximum reach, generate both an image (for Instagram/LinkedIn) and a video (for TikTok/Reels) from the same content.
  1. Preview generated images before uploading. After generating an image with Fireworks, use the Read tool to preview it visually before uploading to Posta. This prevents wasted uploads and media quota.
  1. Use posta_create_post_from_file for multiline captions. Write the caption to a temp file and use the file-based helper instead of trying to embed multiline text in JSON strings. This avoids escaping issues.
  1. Suggest hashtags for posts. When creating a post, suggest 5–10 relevant hashtags based on the caption content, target platform, and topic. Mix broad reach tags (e.g. #AI, #Marketing) with niche tags (e.g. #LaborMarket, #FutureOfWork). Show the suggested hashtags to the user and include them in the post only after the user approves or edits them.
  1. Use /tmp/.posta_last_response for captured output. When capturing posta_api output in a variable with $(), avoid using echo to re-output it — macOS echo corrupts `\

in JSON strings. Instead pipe directly (posta_api ... | jq) or read from the file (jq ... /tmp/.posta_last_response`).

  1. Use platform discovery for validation. Before creating posts for unfamiliar platforms, call posta_get_platform_specs or posta_get_platform "<platform>" to check character limits, required media dimensions, and supported features. This prevents failed posts due to platform constraints.
  1. Auto MIME detection for uploads. When uploading media, you can omit the MIME type parameter — posta_upload_media and posta_upload_from_url auto-detect it from the file content or extension. Only specify MIME type manually when the auto-detection might be wrong (e.g., .bin files).
  1. Always set TikTok privacy level. When creating posts that include TikTok, you MUST include platformConfigurations.tiktok.privacyLevel — TikTok requires it and publishing will fail without it. Use "PUBLIC_TO_EVERYONE" unless the user specifies otherwise. Valid values: PUBLIC_TO_EVERYONE, MUTUAL_FOLLOW_FRIENDS, SELF_ONLY, FOLLOWER_OF_CREATOR.

平台分布

OpenClaw

91.47%
按下载量换算2,342

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills