Automates B2B lead cleaning and classification by analyzing client websites, grading leads A/B/C, updating Feishu tables, and sending reports to Feishu groups.

This skill is not obviously malicious, but exercise caution before installing: - Do not assume embedded credentials belong to you. The package contains hard-coded FEISHU app_id/secret, bitable app_token/table_id, and a chat_id; these could point to someone else's Feishu app and data. Replace them with your own credentials and remove the embedded secrets before deployment. - The SKILL.md says the agent will 'use a browser' to visit company websites, but the shipped code does not fetch or render pages. Verify how website content will be provided to analyzeCompany (the current code expects raw page text). If you need automated crawling, add an explicit, audited crawler and document its permissions. - Confirm the Feishu app permissions and the bitable table schema before giving any credentials; test the skill in an isolated environment or staging account. - Prefer a version of the skill that declares required env vars (FEISHU_APP_ID, FEISHU_APP_SECRET, BITABLE_APP_TOKEN, TABLE_ID, CHAT_ID) instead of embedding secrets. Ask the publisher to clarify why credentials are embedded and to remove them. - If you don't trust the embedded credentials owner, do not allow the skill to run with those defaults — the skill will operate using them unless you override the env vars.

Type: OpenClaw Skill Name: lead-processor Version: 1.0.2 The skill is classified as suspicious due to the presence of hardcoded default Feishu API credentials (`FEISHU_APP_ID`, `FEISHU_APP_SECRET`) within `index.js` and `lead-processor/index.js`. While these can be overridden by environment variables, their inclusion as defaults in the source code represents a minor information security vulnerability. Additionally, the `SKILL.md` instructs the AI agent to use a 'browser tool' to visit external customer websites for 'deep analysis', which, while necessary for the skill's stated purpose, involves a powerful capability that could be exploited if the agent's browsing environment is not adequately sandboxed. The duplication of several files (e.g., `index.js`, `config.json`, `SKILL.md`) in the root and a subdirectory is also an unusual packaging choice, though their content is identical and not malicious.