Security audit
AD, M365, Azure: misconfigurations, inappropriate access, technical review and management review, detailed report.
Cybersecurity
A security posture you can defend.
Active Directory, Microsoft 365 and Azure audits, Zero Trust architecture, Defender and Sentinel, incident response: I raise your security level with clear priorities and measurable results.
Zero Trust architecture
Identities, devices, applications, data: a coherent framework measured against the CISA model.
Detection and response
Defender XDR, Sentinel, SOC integration, incident response plan, alert interpretation.
Active Directory
Secure (tiers, PAW, silos, LAPS) and modernise (hybrid identity, Cloud Sync, gradual decommissioning).
AI agents
Inventory, frame and monitor the agents and connectors acting with your data and identities.
Anti-phishing filtering, SPF, DKIM and DMARC, awareness and simulations for your teams.
What I do for you
Cybersecurity
AD, M365, Azure audit
- Technical review of identified findings
- Management overview with presentation deck
- Detailed report and prioritised action plan
Active Directory: secure and modernise
- Tiering model, admin workstations, authentication silos
- On-premises and cloud restoration plan, incident readiness
- Hybrid identity towards Entra ID and modernisation trajectory
Protecting against AI agents
- Inventory of agents, connectors and related identities
- Scope, least privilege, logging, guardrails
- Detection of uncontrolled usage (shadow AI)
Email security
- SPF, DKIM, DMARC testing and remediation
- Advanced anti-phishing and anti-spoofing filtering
- Awareness and Phishing simulations
Cybersecurity roadmap
- Global strategy and frameworks: NIST, ISO, threat modelling
- Entra, M365, Azure governance; Well-Architected and Cloud Adoption
- CISO support, committees and indicators
Alert interpretation
- Help reading the alerts from your detection tools
- False positive / true positive qualification
- On-demand technical back office
The method, applied
Six steps, one thread.
Containment, hardening, monitoring: the three-phase plan.
What is exposed, what is protected, what remains.
An agent is just one more identity: the same rigour applies.
Method
- 01Scoping
Understand the need, the stakes and the constraints. Set the scope, the actors and the expected results.
- 02Audit
Observe the current state without assumptions: configurations, usage, documents, interviews and tool-based measurements.
- 03Analysis
Confront the findings with frameworks and your objectives. Measure the gaps, qualify the risks, prioritise.
- 04Design
Design the target and the path to reach it: architecture, governance, budget, licences, milestones.
- 05Transcription
Translate the target into operational documents your teams and vendors can execute without me.
- 06Delivery
Present, get stakeholder sign-off, transfer knowledge and support the implementation.
Regain control after AD compromise
Three-phase recovery planCrisis
1 · ContainmentDone
2 · HardeningDone
3 · Monitoring and go-liveIn progress
Isolated bubble · Tiering · PAW · XDR · SIEM
What is our attack surface?
Attack surfaceScore
Illustrative example
Frame the agent that processes our invoices
Agent · Invoice processingFramed
IdentityDedicated identity, passwordless
PermissionsRead invoices, write ERP only
ApprovalHuman above €5,000
LogEvery action traced in the SIEM
Workshops and knowledge transfer
Dense, practical sessions that make your teams autonomous.
Watch
What is moving on this topic
An automatic selection of recent articles from trusted sources, filtered to keep only what touches this specialty.
The CT Conseil briefing
Get this watch every week
One edition a week: the deadlines you must not miss, what is moving on Entra, Purview, Defender and compliance, and one practical tip. No advertising, one-click unsubscribe. Editions are written in French.
Gauge my level in 5 questions
Five questions to gauge your posture. Honesty recommended, nobody is watching.
Two minutes, no tricks. You leave with a level and concrete leads.
Question 1 of 5
AI agents and assistants in your organisation…
Secured? Let's move on to acceleration.
Trusted AI starts where security stops being a brake.