Use Case · MCP Server Security

MCP Servers, Signed and Scoped

MCP servers are the new privileged-dependency class. Your AI agents trust them with production credentials. Safeguard provides the governance layer: verified registry, capability-manifest review, scoped per-server credentials, and full audit logs of every tool invocation.

89+
Verified MCP Tools
100%
Invocations Logged
Per-Server
Scoped Credentials
Signed
Manifest Verification

Why MCP Is npm-2014 All Over Again

Open protocol, explosive ecosystem growth, no governance layer between "someone published it" and "it has your production credentials".

01

Capabilities, Not Just Code

A traditional package dependency is code. An MCP server exposes tools — with capabilities to send email, read files, query databases. The security question is not 'what does the code do?' but 'what permissions does it hand the AI?'

02

Execution Context Is Elevated

An MCP server wrapping your Salesforce API has your Salesforce credentials by design. A malicious server does not need to escape; it was granted the keys at install time.

03

Prompt Injection Becomes RCE-Adjacent

Because tool calls are triggered by LLM output, prompt injection anywhere upstream can trigger tool calls the operator never intended. Every context-window input is effectively a candidate instruction.

04

Stale Servers Accumulate Quietly

Engineers install servers for experiments and never uninstall them. Over months you accumulate ad-hoc tools with stale credentials. None of it is on your asset inventory.

MCP Governance, Enterprise-Ready

Registry. Review. Runtime.

Verified Registry & Manifest Review

Tiered trust model — Tier 1 for official, signed, widely-used servers; Tier 2 for reviewed community servers; Tier 3 blocked in production. Capability manifests reviewed before an MCP server is allowed in the org registry.

Signature verification
Capability manifest audit
Tiered trust rollout

Scoped Credentials Per Server

Purpose-built service identities per MCP server deployment. Credentials scoped to exactly the APIs the server declares needing. No more generic 'MCP runner' role with broad cloud permissions.

Short-lived tokens
Per-server service identity
Capability-scoped permissions

Invocation Audit & Drift Alerts

Every MCP tool call logged with full context: server, tool, arguments, LLM session, user. Drift from declared capability triggers alerts. Anomalous patterns (off-hours, unusual args) surface for review.

Full invocation logs
Capability-drift detection
Integration with your SIEM
Onboarding Flow

From Request To Production In A Reviewable, Auditable Path

A developer requests an MCP server be added. Safeguard automatically fetches the manifest, verifies the signature, pulls down the source for review, and surfaces any dependency, capability, or credential-scope concerns to the security reviewer. On approval, the server is added to the org registry with scoped credentials provisioned and logging enabled. Invocations flow to your SIEM. Capability drift triggers a review ticket. Stale servers expire unless re-affirmed.

5 stages
Request → Deprecate
Per-user
Scoped Credentials
Quarterly
Re-Affirmation
Scenarios

Where This Bites In Real Life

Coding agents and MCP servers in production raise risks SCA was never designed to catch.

01

Coding Agent In A Regulated Repo

An agent has commit rights to a SOC 2 in-scope service. Block writes to production manifests, redact customer IDs in tool output, log every action with the user's identity attached.

02

Multi-Vendor Agent Stack

Engineering uses Claude Code, Cursor, Cline, and a couple of in-house agents. One allowlist, one capability catalogue, one audit trail — regardless of which agent made the call.

03

Prompt-Injection Exfiltration

Adversarial text in a connected vector store tries to trick the agent into leaking credentials. Lion runs inline on every tool response and quarantines the suspicious flow.

04

EU AI Act Logging

Agent activity must be auditable end-to-end with cryptographic provenance. Every tool call, every redaction, every policy decision — signed and chained.

Step By Step

How Safeguard Handles It

01

Agent Connects

The agent hits the Safeguard MCP proxy. Connection metadata — agent vendor, version, environment — is captured before any tool call.

02

Identity & Auth

The agent presents a scoped service identity bound to a human user. No shared bot accounts; every invocation is attributable.

03

Capability Scope Applied

Per-tool capability scope resolved from policy. The agent only sees the tools its role is allowed to invoke.

04

Tool Call Inspected

Arguments scanned for sensitive selectors (production hosts, customer IDs, secrets). Outbound payload checked against the policy bundle.

05

Output Scanned By Lion

Lion runs inline on the response — PII redaction, secret detection, prompt-injection signatures. The agent never sees the raw sensitive bytes.

06

Decision Logged With Chain-Of-Custody

Allow, deny, or transform — the decision is written to the signed audit log with a hash chain back to the previous entry.

07

Response Returned Or Blocked

Clean responses reach the agent; flagged responses return a structured policy error the agent can reason about.

Surfaces

What You See, Ship, And Report

IDE / CLI

In-Editor Approval Prompt

Sensitive tool calls surface a one-click approval prompt inside the IDE. The developer sees the agent's intent, the argument, and the policy verdict before they say yes.

CI / PR

Bot Account Quota & Tokens

Per-agent token usage report attached to the PR. CI checks the quota and flags when an agent is operating above its baseline.

Exec Console

Activity Timeline & Violations

Agent activity timeline across the org, policy-violation rate by tool, and a redacted-output counter that proves Lion is doing the inline work.

Don't Let MCP Outgrow Your Governance.

Put the registry, review process, and audit trail in place before MCP scales across engineering and business teams.