Topic starter
25/10/2026 11:01 am
TL;DR: MCP only pays off when the harness around it turns security questions into trustworthy evidence, according to Corelight, because dynamic retrieval, bounded agent workflows, and drift control matter more than the protocol alone. That shifts the focus from model novelty to governance of traces, permissions, and reviewable behaviour.
Editorial analysis by NHI Mgmt Group, based on content published by Corelight: “What the Black Hat NOC taught me about MCP & agentic SOCs (Chapter 3 of 4)”.
Key questions
Q: How should teams govern AI agents that use MCP?
A: Treat each connected agent as a non-human identity with an owner, a scope, and a review cycle.
Q: Why do self-improving security agents create operational risk?
A: Self-improving agents can drift away from the approved behaviour they started with, especially when changes are driven by traces and no one reviews the resulting code or workflow updates.
Q: What breaks when AI investigations rely only on embedding-based retrieval?
A: Embedding-only retrieval can miss structure, sequence, and windowed change, which are often the real clues in security investigations.
Practitioner guidance
- Define bounded agent roles before production Separate hunting, validation, incident response, reporting, and telemetry-checking duties so each agent has a narrow purpose and minimal authority.
- Treat retrieval design as part of the control plane Use SQL, structured document traversal, and summarisation only where each one matches the investigation question and preserves evidential context.
- Freeze stable agent baselines and review deltas Save known-good versions, require human review for proposed changes, and compare new traces against a rolling baseline before release.
Bottom line: MCP becomes useful in security operations only when the harness around it enforces bounded delegation, evidence capture, and reviewable behaviour.
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →