Bowmark web data: LinkedIn, Amazon, YouTube, Reddit, Airbnb, Instagram
ai.bowmark/bowmark · v8.234.1 · score 89/100 · checked 2026-11-22
Automated, read-only signal from public data, not an audit or an accusation. Maintainers can dispute any finding. How grading works
Web scraping and automation for AI agents: live prices, quotes, bookings, anything behind a login.
Source: https://github.com/bowmark-ai/skill
Findings
| Severity | Rule | Detail |
|---|---|---|
| Low | MW005 | Remote endpoint lists 29 tool(s) to anyone, no credentials needed. |
| Medium | MW102 | 1 tool(s) can execute commands or delete data (delete_connection). Require confirmation before the agent calls them. |
What was checked
- https://api.bowmark.ai/mcp/registry: answers without auth
- pypi: bowmark-mcp@2.1.2 (static checks only; code not executed)
- npm: @bowmark/mcp@2.1.2 (static checks only; code not executed)
Public tools: get_library, run, jig_open, jig_list, jig_search, jig_publish, jig_set_handle, jig_get, jig_update, jig_pull_update, jig_read, jig_look, session_call, session_close, report, get_usage, llm_poll, llm_answer, llm_detach, support, list_secrets, request_secret, get_secret_link, list_connections, add_connection, wait_for_sign_in, lease_ip, delete_connection, logout_connection
Maintainer? Add the badge
Grades refresh daily. Think a finding is wrong? Dispute this grade.
[](https://www.mcpwatch.io/s/ai-bowmark-bowmark-60b719ed)