mcpwatch

← All servers

Grade D

Garita MCP

io.github.Garitalab5/garita-mcp · v1.0.0 · score 57/100 · 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

30 security tools + the Vigia agent to audit MCP servers before you connect.

Source: https://github.com/Garitalab5/garita-mcp-server

Findings

SeverityRuleDetail
LowMW005Remote endpoint lists 31 tool(s) to anyone, no credentials needed.
CriticalMW101tool 'detect_poisoned_tool_descriptions': description contains instruction-like text aimed at the model ('<IMPORTANT>'). Possible tool poisoning.

What was checked

Public tools: audit_mcp_server, simulate_approval, detect_poisoned_tool_descriptions, check_known_vulnerabilities, detect_tool_rug_pull, detect_poisoned_config_files, check_package_authenticity, detect_memory_poisoning, detect_ssrf_risk, detect_path_traversal_risk, detect_command_injection_risk, detect_sql_injection_risk, detect_token_passthrough_risk, detect_hardcoded_secrets_risk, detect_insecure_deserialization_risk, detect_toxic_flow_risk, detect_ssti_risk, detect_dynamic_code_risk, detect_dns_rebinding_risk, detect_missing_auth_risk, detect_mcp_stdio_launch_risk, detect_redos_risk, detect_session_hijack_risk, detect_oauth_audience_confusion_risk, detect_unbounded_allocation_risk, detect_cross_client_leak_risk, detect_sampling_hijack_risk, detect_output_schema_ssrf_risk, detect_excessive_agency_risk, detect_call_spoofing_risk

Maintainer? Add the badge

Grades refresh daily. Think a finding is wrong? Dispute this grade.

[![mcpwatch](https://www.mcpwatch.io/badge/io-github-garitalab5-garita-mcp-7585c76a.svg)](https://www.mcpwatch.io/s/io-github-garitalab5-garita-mcp-7585c76a)