MCP security report
mcp.agentspodium.com
F34/100
Security grade FA few things are worth a closer look before connecting.
Connected ✓13 tools scanned
https://mcp.agentspodium.com/mcp
mediumlist_instances
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumcreate_instance
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumset_llm_key
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumset_webhook
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumset_peers
Configures inter-agent trust and token exchange without explicit safety guardrails
This tool lets an instance be configured to call arbitrary external A2A endpoints and to store bearer tokens for them. If a malicious or compromised URL is added (e.g. via prompt injection elsewhere), the instance could send its context/data to attacker-controlled endpoints along with a token. There's no mention of URL validation or restrictions here (unlike set_webhook's explicit loopback/private-host refusal), which is inconsistent and could allow SSRF-style misuse via peer URLs pointing to internal services.
low
Not on gate's verified list
This server isn't on gate's handpicked list. That's not necessarily bad, but there's no third-party signal about it — review its tools carefully.
lowset_llm_key
LLM provider API key transmitted and stored by third-party service
The tool collects a raw LLM provider API key (e.g. OpenAI/Anthropic key) as a plaintext parameter to be stored on the AgentsPodium platform. While described as 'stored encrypted, never echoed back', the agent should ensure the user is aware they are handing a live credential to a third-party hosting service, and that this key could be used to run up charges or access other resources tied to that key. Not malicious per se, but a sensitive-credential handling point worth flagging to users before invocation.
lowcreate_instance
extraSoul field allows injecting arbitrary system-prompt instructions
The extraSoul parameter appends up to 4000 characters of free-form 'system-prompt instructions' to the persona. This is a legitimate customization feature, but since it directly shapes the deployed agent's behavior, an agent operator should treat content passed here as sensitive/instructional and avoid blindly forwarding untrusted user text into it, as it could be used to embed hidden or manipulative instructions into the resulting instance's system prompt.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/3586d65e-f289-4f54-8e90-514f355125c7)Scanned 9/12/2026 · This is a point-in-time snapshot of the server’s public tools. Re-scan.
Connect this server safely with gate.
Continuous re-scans, drift alerts, per-tool allow/ask/block, and a log of every call.
Monitor it — free