PromptGuard
co.promptguard/security · v1.2.5 · 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
Scan prompts for injection attacks, redact PII, and audit LLM SDK usage from any MCP client
Source: https://github.com/promptguardhq/promptguard
Findings
| Severity | Rule | Detail |
|---|---|---|
| Low | MW005 | Remote endpoint lists 13 tool(s) to anyone, no credentials needed. |
| Medium | MW002 | Package ghcr.io/promptguardhq/promptguard-mcp-server:1.2.5 is published without a fixed version. |
What was checked
- https://api.promptguard.co/mcp: answers without auth
- npm: @promptguard/mcp-server@1.2.5 (static checks only; code not executed)
- oci: ghcr.io/promptguardhq/promptguard-mcp-server:1.2.5@? (static checks only; code not executed)
Public tools: promptguard_scan_text, promptguard_scan_code, promptguard_redact, promptguard_status, promptguard_list_events, promptguard_get_usage, promptguard_list_policies, promptguard_list_audit_log, promptguard_list_projects, promptguard_get_fleet_status, promptguard_list_red_team_tests, promptguard_list_webhook_deliveries, promptguard_propose_suppression
Maintainer? Add the badge
Grades refresh daily. Think a finding is wrong? Dispute this grade.
[](https://www.mcpwatch.io/s/co-promptguard-security-ffc59bdb)