n
AI
vigate
News
Events
Models
Model Finder
Which model for your need, in 6 questions
Comparator
Filters, side by side, verified data
Model cards
Price, licence, hardware, deploy
Guides
Which model for which need
Podium
This month’s ranking
Baromètre
The market in figures, monthly
Trust Index
Enterprise trust per model
Hardware Match
Which model for my machine
API & MCP
The data in your agents
Tools
Tool Finder
Which tool for your need, in 4 questions
My AI toolbox
Several tasks + a budget → a coherent, shareable kit
Practical recipes
One task, the right tools, the steps, tested
AI visibility checker
Can ChatGPT, Claude, Perplexity read your site? Score in 10 s
Claude Code checker
Is your configuration safe? Score and corrected file
By use
Photo, video, audio, writing, code, everyday
Run your AIs
Local, orchestration, routers, memory, security
MCP servers
Connectors for your agents, scored
Tools
The actions your AI can call
Skills
The methods your agents follow
Plugins
Claude Code, Chrome, Obsidian, Office, Figma…
Monthly edition
The best, the new, the alerts
Learn
Articles
Deep dives, how-tos, analysis
Cheat sheets
The essentials in 90 seconds
Glossary
50 terms, 8 categories
About
Studio
The journey
Audit, implementation, training
Studio offers
AI Radar, Sprint, Training, Steering
Free AI diagnostic
5 minutes, instant result
FR
/
EN
LIVE
Anthropic says Claude helped trace the origin of NIST's mysterious ECDSA seeds
16/10/26 · Anthropic
|
Reflection AI unveils Beam, a 501-billion-parameter open model
16/10/26 · Reflection AI
|
Meta releases Muse Spark Open, a mixture-of-experts model built for local Mac Studio inference
15/10/26 · Meta
|
Meta releases Muse Spark Open, an open-weight multimodal model
15/10/26 · Meta
|
Google DeepMind outlines safety measures behind Gemini 4 Argon
13/10/26 · Google DeepMind
|
Tev-Gym: A Training Recipe to Fix Sparse Rewards in Long-Horizon Agents
13/10/26
|
Google DeepMind references Gemini 4, code-named Argon
13/10/26 · Google DeepMind
|
Google DeepMind unveils Gemini 4 Argon, a robotics-focused variant
13/10/26 · Google DeepMind
|
Character.AI files for Chapter 11 bankruptcy, sells assets to Disney
13/10/26 · Character.AI
|
Study Shows AI Models Can Provably Conceal Arbitrary Capabilities
13/10/26
|
Websites are increasingly shutting out AI agents
18/10/26
|
NVIDIA RTX Spark laptops bring local AI to premium Windows PCs
17/10/26 · NVIDIA
|
Anthropic says Claude helped trace the origin of NIST's mysterious ECDSA seeds
16/10/26 · Anthropic
|
Reflection AI unveils Beam, a 501-billion-parameter open model
16/10/26 · Reflection AI
|
Meta releases Muse Spark Open, a mixture-of-experts model built for local Mac Studio inference
15/10/26 · Meta
|
Meta releases Muse Spark Open, an open-weight multimodal model
15/10/26 · Meta
|
Google DeepMind outlines safety measures behind Gemini 4 Argon
13/10/26 · Google DeepMind
|
Tev-Gym: A Training Recipe to Fix Sparse Rewards in Long-Horizon Agents
13/10/26
|
Google DeepMind references Gemini 4, code-named Argon
13/10/26 · Google DeepMind
|
Google DeepMind unveils Gemini 4 Argon, a robotics-focused variant
13/10/26 · Google DeepMind
|
Character.AI files for Chapter 11 bankruptcy, sells assets to Disney
13/10/26 · Character.AI
|
Study Shows AI Models Can Provably Conceal Arbitrary Capabilities
13/10/26
|
Websites are increasingly shutting out AI agents
18/10/26
|
NVIDIA RTX Spark laptops bring local AI to premium Windows PCs
17/10/26 · NVIDIA
|
+ nAIvigate
SHEETS FOR PEOPLE WITHOUT 2 HOURS
+ nAIvigate ·
fiches/
▣ index.md
Cheat
Sheets.
One AI topic = one sheet.
Dense, actionable, always current.
Scannable in 90 seconds.
#008
Agentic pentesting: same model, opposite results.
Tools are multiplying and all promise critical findings. Six questions, a scoring grid and eight red flags to choose on evidence.
SECURITE
3 MIN
#007
Claude Code now starts in auto mode. Who decided that in your company?
Since version 2.1.283, sessions open without permission prompts, under a classifier's review. The six modes, the enterprise settings and the mode to enforce in each context.
SECURITE
2 MIN
#006
CCA-F: 15 days of revision? One page first.
Claude Certified Architect – Foundations (CCAR-F): verified specs, the 5 domains by weight, the 10 reflexes the exam expects, the recurring distractors, and a D-7 plan.
CERTIFICATION
2 MIN
#005
Skills & MCP: the fiche that maps the whole path
The decision-ready digest of the 9-step path
FONDAMENTAUX
2 MIN
#004
The 50 AI terms to master for 2026 decisions
Not another glossary: for each term, the definition, the decision it implies, and the costly pitfall. Built for CIOs/CISOs, readable by any exec.
LEXIQUE
12 MIN
#003
AI agents: your new employees have no badge.
They read your mail, write to your databases, push to your repos, with someone else’s credentials. The risk map, frameworks and actions.
SECURITE
2 MIN
#002
The 40 things to install in Claude.
Harness, memory, tools, security, cost: the 39 repos that matter, stars verified live on GitHub.
OUTILS
2 MIN
#001
AI Act: postponed? Not the way you think.
What actually applies since August 2, 2026 (penalties included) and what the omnibus really moved.
REGLEMENTAIRE
2 MIN
⎇ main
8 sheets published
naivigate.com/fiches