Daily Market Insight
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill matches its daily market-report purpose, but it is set to automatically create Feishu documents and post to a group with unclear destination and approval scope.
Install or enable this only if you are comfortable with a daily automated market report being created and posted in Feishu. Confirm the exact Feishu account, workspace, document destination, and chat, and consider requiring manual approval before each scheduled post. Also verify the local opencode/sisyphus setup and treat the generated market predictions as informational, not financial advice.
Findings (4)
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.
It could create documents and send notifications in a Feishu group every day, potentially in the wrong chat or before a user reviews the report.
The skill is designed to run automatically and write/send into Feishu, but the exact destination and approval boundary are not specified.
schedule: "0 10 * * *" ... 使用 `feishu_doc` 创建新文档 ... 在当前飞书群发送消息
Make the Feishu workspace, document location, and chat explicit, and add a manual approval or dry-run mode before posting.
The skill may act through the user's or workspace's Feishu permissions to create documents and send messages.
Using Feishu documents and group messaging implies delegated workspace/account access. This is expected for the stated purpose, but the artifact does not define the credential or permission scope.
requires: ["web_search", "web_fetch", "feishu_doc"] ... 使用 `feishu_doc` 创建新文档
Confirm which Feishu account/app is used and restrict permissions to the intended workspace, document folder, and chat.
Report quality and local behavior depend on the installed opencode tool and the sisyphus agent configuration.
The workflow invokes a local command and an external/local agent. This is aligned with the multi-agent analysis purpose, but the binary and agent provenance are not provided in the skill artifacts.
opencode run --agent sisyphus "分析以下财经新闻..." ... 先在目标目录执行 `git init`
Use a trusted opencode installation, review the sisyphus agent configuration, and run it in a dedicated project directory.
A malicious or low-quality web page could influence the generated report if the agent over-trusts fetched content.
External web pages are fetched and then passed into an analysis agent. This is purpose-aligned, but fetched pages can contain untrusted text that should be treated only as data.
使用 `web_search` 搜索 ... 使用 `web_fetch` 获取详细内容 ... 调用 **opencode sisyphus** 进行分析
Treat fetched web content as untrusted, cite sources, and avoid publishing automatically without review.