← 返回 Skills 市场
rotbit

Creator Alpha Feed

作者 rotbit · GitHub ↗ · v1.0.8
cross-platform ⚠ suspicious
1720
总下载
0
收藏
4
当前安装
9
版本数
在 OpenClaw 中安装
/install creator-alpha-feed
功能描述
Collect and rank daily AI content for creator-focused publishing workflows. Use when users ask for AI topic scouting, KOL tracking (especially X/Twitter), practical tutorial picks, industry updates, or automated Feishu/Obsidian briefing pushes with configurable templates and time windows.
使用说明 (SKILL.md)

Creator Alpha Feed

  1. Read config first:
    • ${OBSIDIAN_CONFIG_PATH:-\x3Cyour_obsidian_vault>/OpenClaw/项目/AI内容日报/采集配置.md}
  2. Execute collection in this order for X:
    • homepage feed → whitelist accounts → keywords
  3. Prefer API where available; fallback to browser when unavailable.
  4. Enforce browser tab cap:
    • max 7 concurrent tabs; close finished tabs first; end with 0 tabs (close all temporary tabs before finishing).
  5. Build ranked outputs by configured structure (default):
    • KOL TOP3 (last 6h)
    • Practical/Tutorial/Opinion TOP10
    • Industry TOP3 (last 6h)
  6. Push concise results to group channel; write full report to Obsidian path.
  7. Name report files with timestamp format: YYYY-MM-DD_HHMM.md.
  8. Prefer real Obsidian Vault path (not workspace mirror) when available.
  9. Use structured Obsidian directories:
    • OpenClaw/项目/AI内容日报/01-日报/ for final reports
    • OpenClaw/项目/AI内容日报/02-运行记录/ for verification/debug runs
    • OpenClaw/项目/AI内容日报/03-文档/ for installation/operational docs
  10. If login is required for a source, pause and notify user to log in; wait up to 3 minutes with periodic checks, then continue remaining sources if still unavailable.

Bundled scripts

Use scripts/collect-v4.sh and related scripts for deterministic fallback/automation when needed.

Required output checks

  • Include must-track account status for @xiaohu @dotey @marclou
  • Include fallback/degradation notes
  • Include final report path
  • In group replies, mention the question asker (@who asked) when channel supports mentions
安全使用建议
This skill appears to implement the described content-collection and reporting workflow, but exercise caution before running it: - The skill's metadata declares no environment variables, yet SKILL.md and the scripts expect environment variables and local paths (Obsidian vault paths, WORKSPACE_DIR, FEISHU_CHAT_ID/FEISHU_USER). Before installing or running, confirm which environment variables and credentials you will supply and why. - Review the bundled scripts yourself (they are plain shell) — they make many network requests (hn.algolia, reddit, TechCrunch via rss2json, rsshub.app, multiple Nitter instances, etc.) and will write files into your home (~/.openclaw/workspace) and into the Obsidian vault path you provide. If you don't want external network access or writes into your vault, don't run them. - If you plan to enable Feishu pushes, supply a dedicated chat/token with minimal scope; do not reuse high-privilege tokens. The scripts reference FEISHU_CHAT_ID/FEISHU_USER but do not handle token storage — determine how OpenClaw will authenticate pushes before providing credentials. - Run in a sandbox or with restricted test vault/workspace first. Validate outputs and that no unexpected endpoints are contacted (check logs/collection.log). Monitor what the agent sends to sessions_spawn or other LLM calls — collected raw content is fed into AI analysis and could be transmitted externally. - If you cannot or will not audit the scripts, avoid giving the skill write access to your real Obsidian vault or production workspace; instead provide a dedicated test folder and test credentials. Given the undeclared env/config requirements and filesystem/network footprint, treat this skill as potentially risky until you confirm and control the environment it will run in.
功能分析
Type: OpenClaw Skill Name: creator-alpha-feed Version: 1.0.8 The skill is classified as suspicious due to its extensive use of high-risk capabilities, specifically network access via `curl` to various external domains (e.g., `hn.algolia.com`, `www.reddit.com`, `api.rss2json.com`, `nitter.net`, `rsshub.app`, `x.com`) and explicit instructions for the OpenClaw agent to use its `browser` tool (`browser open`, `browser snapshot`). While these actions are plausibly needed for the stated purpose of collecting AI content, browser automation inherently carries risks. However, the skill includes explicit instructions in `scripts/twitter-browser-tasks.sh` and `references/data-collection-rules.md` to limit browser access to public pages only and prohibit collecting private information, indicating an intent to mitigate these risks rather than exploit them.
能力评估
Purpose & Capability
Name/description (AI content scouting, KOL/Twitter tracking, Obsidian/Feishu pushes) matches the included scripts: multiple collection scripts, ranking/analysis, and push guidance. However the skill metadata declares no required env vars or config paths while SKILL.md and scripts reference several environment variables and local paths (OBSIDIAN_CONFIG_PATH, OBSDIAN_REPORT_DIR, WORKSPACE_DIR, FEISHU_CHAT_ID/FEISHU_USER). That is an incoherence: the skill expects credentials/config but does not declare them.
Instruction Scope
SKILL.md instructs reading/writing real Obsidian vault paths, writing reports into specific directories, using OpenClaw browser/web_fetch and sessions_spawn tooling, and to pause for user login when needed. The bundled scripts perform wide web fetching (hn.algolia, reddit, TechCrunch via rss2json, rsshub.app, multiple Nitter instances) and write files under ~/.openclaw/workspace or the user's Obsidian vault. These actions are within the stated purpose but operate on local files and many remote endpoints — the instructions grant broad discretion to access web content and to write into user-owned directories that were not declared as required/configured in metadata.
Install Mechanism
No install spec (instruction-only in registry), which reduces distribution risk. However the skill bundle includes many shell scripts that will be present and runnable; they use curl/jq and rely on platform tooling (openclaw browser, sessions_spawn). There are no downloads from untrusted URLs nor archive extraction in the install phase, which is good, but the presence of executable scripts means executing them will perform network I/O and file writes.
Credentials
Although the registry lists no required environment variables or primary credential, the SKILL.md and scripts reference several environment variables (OBSIDIAN_CONFIG_PATH, OBSDIAN_REPORT_DIR, WORKSPACE_DIR, FEISHU_CHAT_ID, FEISHU_USER) and expect the agent to be able to send messages to Feishu and write into an Obsidian vault. Requesting or using chat IDs / messaging tokens and direct filesystem paths is proportionate to the stated push-to-Feishu / write-to-Obsidian functionality — but the fact these are not declared in the metadata is an inconsistency and makes it unclear what credentials the skill will need or try to use.
Persistence & Privilege
always is false and the skill does not request to be force-enabled. The scripts write to their own pipeline directories (e.g., ~/.openclaw/workspace/ai-content-pipeline) and to user-provided Obsidian paths; they do not modify other skills' configs or request elevated system-wide privileges. No evidence of automatic long-term persistence beyond normal files created under the user's workspace/vault.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install creator-alpha-feed
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /creator-alpha-feed 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.8
Add structured Obsidian directory convention (01-日报/02-运行记录/03-文档) and enforce report placement + timestamp naming.
v1.0.7
Stabilize latest production flow: core-template output, blogger-focused prioritization, login-wait handling, and Obsidian timestamp naming (YYYY-MM-DD_HHMM).
v1.0.6
Add login-handling flow: prompt user to log in, wait with retries, then continue other sources without breaking full run.
v1.0.5
Update: enforce Obsidian report filename format YYYY-MM-DD_HHMM.md and prefer real Obsidian Vault path over workspace mirror.
v1.0.4
Fix browser cleanup: enforce closing all tabs after collection (0 tabs at end).
v1.0.3
Security hardening: removed logged-in/home timeline collection guidance; switched to public X pages only; added explicit privacy-safe collection constraints.
v1.0.2
Privacy scrub: removed machine/user-specific identifiers and absolute local paths; replaced with env placeholders and generic config vars.
v1.0.1
Add mention rule: @ the asker in group replies; keep 3-track briefing workflow and output checks.
v1.0.0
Initial release: 3-track AI content workflow (KOL TOP3@6h + Practical TOP10 + Industry TOP3@6h), configurable via Obsidian + Feishu push.
元数据
Slug creator-alpha-feed
版本 1.0.8
许可证
累计安装 5
当前安装数 4
历史版本数 9
常见问题

Creator Alpha Feed 是什么?

Collect and rank daily AI content for creator-focused publishing workflows. Use when users ask for AI topic scouting, KOL tracking (especially X/Twitter), practical tutorial picks, industry updates, or automated Feishu/Obsidian briefing pushes with configurable templates and time windows. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1720 次。

如何安装 Creator Alpha Feed?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install creator-alpha-feed」即可一键安装,无需额外配置。

Creator Alpha Feed 是免费的吗?

是的,Creator Alpha Feed 完全免费(开源免费),可自由下载、安装和使用。

Creator Alpha Feed 支持哪些平台?

Creator Alpha Feed 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Creator Alpha Feed?

由 rotbit(@rotbit)开发并维护,当前版本 v1.0.8。

💬 留言讨论