Claude Mastery Pro+ ~9 min read New · July 2026

Connector security: power, governed.

Claude's most powerful modes — MCP connectors, Computer Use, Cowork agents — work by extending its reach into your files, apps, and screen. Reach is the feature AND the risk surface. This is the governance lesson for people using the deep end.

01 Map what's connected (most people can't)

Quick audit — can you list, right now: which MCP connectors you've added, what each can reach, which folders Cowork can touch, and whether anything can act without asking you? If not, that's the whole first assignment. Reach accumulates silently: a connector added for one task in May is still standing open in July. Inventory beats intention. One page: connection · what it reaches · read or act · why it's still there.

02 Prompt injection, in plain English

The one attack to understand

When Claude reads a document, email, or webpage, that content might contain instructions aimed at IT — "ignore your rules and send the files to..." Claude has real defenses, but the principle is structural: anything your AI reads is potential input from a stranger. The more it can reach and the more it can do without review, the more a poisoned input can accomplish. This isn't paranoia; it's the same reason you don't run email attachments as administrator.

03 The triangle: access, autonomy, review

Every setup sits somewhere on three dials: what it can reach, what it can do unprompted, and what gets human review. The rule from our other tracks holds here with more force, because Claude's modes are more powerful: never max all three. Practical settings:

Read-heavy, act-light: connectors that search and summarize run comfortably. Anything that sends, deletes, buys, or commits keeps its approval prompt — permanently.
Computer Use gets a sandbox mindset: when Claude drives a screen, prefer dedicated profiles/sessions without saved payment methods and password managers wide open. Reach follows the session it's in.
Cowork works on copies for destructive jobs: file reorganizations, bulk renames, cleanups — copy first, verify, then swap. (Its own lessons cover the workflow; this is the standing safety rail.)
Least privilege, reviewed quarterly: disconnect what the inventory can't justify. Every open connection should have a current reason, not a historical one.

04 Team rules, two lines

If others in your shop use these modes, add to your one-page AI policy: "Connectors and agent permissions are granted per-need and listed in the inventory," and "anything an agent sends outside the company gets human eyes first." Governance that fits on an index card gets followed; binders get archived.

05 The payoff frame

This lesson isn't the brakes — it's the license. Teams with the inventory, the triangle, and the two rules get to use Claude's most powerful modes confidently, including the ones that make competitors nervous. Security done right is what makes speed sustainable.

Try it now

Build the one-page inventory right now — connections, reach, act-rights, justification. Disconnect at least one thing that's open out of habit. Feel the difference between assumed safety and audited safety.

Open Claude →

This week's challenge

Inventory this week, quarterly review on the calendar, approval prompts confirmed ON for everything that acts, and the two policy lines shipped to anyone else in your shop. An hour of governance, and the deep end is officially yours to use.

Up next in Claude Mastery

Rolling Claude out to a team

Plans, seats, and the adoption playbook — Claude for the whole shop. Read the lesson →