← 返回 Skills 市场
458
总下载
0
收藏
2
当前安装
16
版本数
在 OpenClaw 中安装
/install purefeed
功能描述
Monitors Twitter/X feeds with AI signal detection. Searches tweets semantically, manages signal detectors, generates human-sounding posts, checks AI detectio...
安全使用建议
This skill appears coherent: it uses purefeed.ai endpoints and only needs your Purefeed API key and curl. Before installing, confirm you control the Purefeed account and understand that publishing flows will post to X if you have connected Typefully — review and restrict that integration if you don't want automated posts. Because the skill's source/homepage is not provided, consider: (1) obtain the API key only from your Purefeed account page, (2) use a limited/rotatable API key, (3) verify Purefeed account settings and what permissions Typefully has, and (4) avoid granting the skill access to other unrelated credentials. If you want extra caution, test read-only operations first (GET /auth/me, GET /styles, GET /signals) before allowing any publish actions.
功能分析
Type: OpenClaw Skill
Name: purefeed
Version: 0.10.0
The purefeed skill bundle is a legitimate integration for the Purefeed AI service, designed to monitor Twitter/X feeds and automate content publishing via the purefeed.ai API. It utilizes bash and curl to interact with documented endpoints for signal detection, tweet searching, and post generation. The instructions in SKILL.md are strictly functional, providing workflows and formatting rules consistent with the stated purpose, with no evidence of data exfiltration, malicious execution, or harmful prompt injection.
能力评估
Purpose & Capability
Name/description (monitoring, semantic search, draft/publish) align with the declared requirements: only curl and PUREFEED_API_KEY are required. Asking for an API key to call purefeed.ai is expected for this functionality.
Instruction Scope
SKILL.md contains explicit API call sequences (signals, feed, styles, studio endpoints) and concrete error handling. It does not instruct reading unrelated files or arbitrary environment variables. It does direct publishing workflows (via Typefully) — expected given the stated publish capability.
Install Mechanism
Instruction-only skill with no install spec and no code files; this is the lowest-risk install model. It relies on curl already present on the environment.
Credentials
Only a single credential (PUREFEED_API_KEY) is required and declared as primaryEnv. Other keys mentioned (LLM API keys, Typefully connection) are part of the Purefeed account configuration, not additional env vars requested by the skill. The credential request is proportional to the service it integrates with.
Persistence & Privilege
always is false and the skill does not request system-wide config changes. The README suggests setting the skill's own env in OpenClaw config (normal). It can publish content if the user's Purefeed account is connected to Typefully — this is an expected capability, not a hidden elevation.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install purefeed - 安装完成后，直接呼叫该 Skill 的名称或使用
/purefeed触发 - 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
v0.10.0
Fix signal creation, improve onboarding UX, auth error handling
v0.9.0
Add folder management endpoints: create, rename, delete folders and add/remove tweets from folders
v1.1.0
Fix: don't auto-include keywords (too restrictive), add color support, always set tags
v1.0.0
Major restructure: lean SKILL.md (145 lines) + separate API_REFERENCE.md for progressive disclosure. Stronger output formatting rules. Better LLM compliance.
v0.8.0
Move formatting rules to top of skill for better LLM compliance
v0.7.0
Add formatting rules for clickable @mentions
v0.6.0
Fix signal creation (auto-activate, ownership, cron/timezone), add schedule management to PUT /signals/:id, fix response shapes, correct feed response structure, use FTS-only search
v0.5.1
No changes detected in this release.
- Version bump only; no file or functionality changes.
v0.5.0
- Add signal filtering to GET /signals: new query params `search`, `active`, and `tags`.
- Document semantic search on signals via `search` param.
- Show how to list only active signals matching search or tags.
- Update GET /signals examples to demonstrate new filters.
v0.4.2
- Documentation only: SKILL.md formatting changes.
- No functional or API changes included.
v0.4.1
- Documentation updated in SKILL.md; no functional or API changes.
- Improved clarity and formatting of existing instructions and endpoint descriptions.
v0.4.0
- Expanded and clarified the "Signals" documentation: now details active/inactive status based on the signals_subscriptions array and response filtering.
- Improved API Reference with explicit REST paths/methods, parameter breakdowns, and detailed response fields for each endpoint.
- Updated Feed, Signals, and Folders sections to match current API structure, including changes to filters and returned fields.
- Removed redundant details, clarified tool dependencies, and reorganized examples for easier usage.
- Adjusted content to reflect up-to-date features and usage rules.
v0.3.2
- Updated API base URLs from `https://purefeed.ai` to `https://www.purefeed.ai`
- Removed requirement for `jq` binary; only `curl` is now required
- Tweaked authentication and example responses for improved clarity
- Minor formatting and consistency changes to code examples and endpoint documentation
v0.3.0
**purefeed 0.3.0**
- Major SKILL.md rewrite: now focused on a concise, shell-command–centered API quick-reference.
- Adds explicit OpenClaw/CLI setup instructions and environment variable requirements.
- Lists all API endpoints with bash `curl` examples for each core function.
- Specifies tool dependencies, error handling, and main usage flows in a simplified, technical format.
- Adds metadata for OpenClaw compatibility, including required binaries and environment config.
- Removes beta/discussion sections, replacing them with direct API doc structure for faster onboarding.
v0.2.0
Add OpenClaw setup instructions, fix package name to @purefeed/mcp-server, add beta warning
v0.1.0
Initial beta release — AI-powered Twitter/X feed intelligence
元数据
常见问题
Purefeed 是什么？
Monitors Twitter/X feeds with AI signal detection. Searches tweets semantically, manages signal detectors, generates human-sounding posts, checks AI detectio... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 458 次。
如何安装 Purefeed？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install purefeed」即可一键安装，无需额外配置。
Purefeed 是免费的吗？
是的，Purefeed 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Purefeed 支持哪些平台？
Purefeed 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 Purefeed？
由 poloniki（@poloniki）开发并维护，当前版本 v0.10.0。
推荐 Skills