fastfish 微信公众号(wechat)快速排版精简版

PassAudited by ClawScan on May 1, 2026.

Overview

The skill is coherent and disclosed, but users should notice that it guides installation of an external GitHub service, uses optional webhook/API credentials, and can create scheduled hotspot-push jobs.

Before installing, review the linked GitHub repository and dependencies, use the pinned release tag, keep credentials in .env, and only enable cron/webhook push jobs for channels you control and want to receive ongoing hotspot messages.

Findings (5)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Installing the skill’s service means trusting the referenced repository and its Python dependencies.

Why it was flagged

The skill tells users to install and run code from an external GitHub repository. This is disclosed and includes mitigation advice, but it is still a supply-chain point users should review.

Skill content
本 Skill 会指导安装并运行来自 GitHub 的第三方仓库。**供应链风险**:clone + pip install 会执行外部代码
Recommendation

Review the repository and dependencies, use the pinned release tag as recommended, and install in an isolated non-root environment.

What this means

The agent may run local commands for the service when the user asks for formatting, hotspot retrieval, setup, or preview actions.

Why it was flagged

The skill uses system.run to execute local Python scripts. The scope is bounded to named scripts and arbitrary commands are prohibited, making this purpose-aligned but still worth noticing.

Skill content
**3. system.run 仅执行本 Skill 文档列出的脚本** ... 允许:`fastfish_cli.py`、`get_hot_now.py`、`push_hot_to_im.py`、`fetch_hot_items.py`、`init_hot_push_config.py`、`main.py`
Recommendation

Only invoke actions you intend, and confirm commands are limited to the documented scripts.

What this means

Webhook URLs, bot tokens, chat IDs, and API keys can authorize messages or service access if mishandled.

Why it was flagged

The skill may use API keys and messaging webhook/token credentials. These credentials are purpose-aligned for API authentication and hotspot push delivery, and the document instructs not to expose .env values.

Skill content
MEDIA_AGENT_API_KEY (可选,API 鉴权);热点推送至少其一:HOT_PUSH_FEISHU_WEBHOOK, HOT_PUSH_DINGTALK_WEBHOOK, HOT_PUSH_DINGTALK_SECRET(钉钉加签), HOT_PUSH_TELEGRAM_BOT_TOKEN+CHAT_ID,存 .env
Recommendation

Use least-privilege credentials, store them only in .env as described, and rotate them if they are ever shown or committed.

What this means

Content sent through configured channels leaves the local environment and is visible to the selected chat, channel, or workspace.

Why it was flagged

The skill can send hotspot summaries through external messaging channels and webhooks. This is disclosed and user-configured, but users should verify destinations before enabling pushes.

Skill content
推送支持飞书/钉钉/Telegram(系统 crontab 或 OpenClaw Cron 执行 push_hot_to_im.py)及 Slack/Discord 等(OpenClaw Cron announce)。
Recommendation

Confirm the webhook/channel destination and avoid sending private drafts or sensitive content through public or shared channels.

What this means

Once configured, scheduled jobs may continue sending hotspot messages automatically until removed or disabled.

Why it was flagged

The skill documents creation of scheduled jobs that continue running after setup. This persistence is central to daily hotspot push and is accompanied by rules against unrequested cron edits.

Skill content
openclaw cron add --name "每日热点" --cron "0 8 * * *" --tz "Asia/Shanghai" --session isolated --message "cd /opt/fastfish-lite && python scripts/push_hot_to_im.py,将热点推送到配置的渠道"
Recommendation

Create cron jobs only when you want ongoing pushes, record the job ID, and remove or disable the schedule when no longer needed.