The real workflow
Where Claude Code enters the work
The agent workflow can combine repository reading, file edits, terminal commands, dependency installation, tests, and network access.
Claude Code is a local or cloud coding agent with file, command, network, MCP, and editing capabilities governed by permissions, sandboxing, trust, and account settings.
MCP servers and hooks can expose local commands, files, APIs, tokens, and external services to the agent.
Claude Code only has the permissions granted to it, but broad read access, bypass modes, unsandboxed commands, or overpowered MCP servers can make that boundary much wider than expected.
The presence of this path does not prove an incident. It identifies the boundary that should be checked before more sensitive context or authority is added.
Tool-specific boundary
Inspect the real access points.
What may carry context
repository and local file reads
edits and Bash commands
network access, MCP servers, hooks, and cloud environments
Settings to verify
Permission mode and deny rules
Filesystem and network sandbox
Trusted directories, MCP servers, hooks, and unsandboxed escape paths
Why this context matters
The consequence for developers
Developer workflows join high-value source code with tools that can retrieve context, propose changes, run commands, and cross trust boundaries quickly. In this case, a business connector can turn an over-privileged account into a broad retrieval or action surface spanning customers, employees, projects, and internal operations.
A connector does not create data, but it can make existing account permissions available through a new interface. The safe question is not only whether the connector is trusted; it is whether the connected account is broader than the task requires.
The team can reproduce what the tool accessed, separate read and write authority, protect secrets, and review consequential changes before execution.
Context decision
Three questions before adding access
What can this session read, write, execute, contact over the network, and approve without another person?
Are secrets, production data, protected branches, deployment credentials, and unrelated repositories outside the effective scope?
Will the final diff, commands, dependency changes, test evidence, and approvals survive after the session closes?
Evidence goal: Produce a reproducible technical record of roots, permissions, denied paths, network policy, generated changes, approvals, tests, and rollback points.
A repeatable review
Four steps, no sensitive data required
- 1
Write down the exact Claude Code account, workspace, project, device, and connected service used in this workflow.
- 2
Review .mcp.json, managed settings, hooks, server commands, scopes, transports, and trust prompts.
- 3
Assign the decision and next review to the repository owner or engineering lead; do not leave the access boundary as an unwritten assumption.
- 4
Remove unknown servers and sandbox each remaining integration with narrow paths and network destinations. Record the result without copying private content or raw credentials into the report.
Controls to apply
Reduce access before adding trust
Remove unknown servers and sandbox each remaining integration with narrow paths and network destinations.
Use a least-privilege account or service identity created for the specific workflow.
Separate read-only retrieval from write, send, share, delete, and financial actions.
Set a recurring owner and expiry date for every connector rather than leaving access permanent.
Decision rule
Know when a formal baseline is justified
If the assistant has no connectors, document that and keep it true. If it can retrieve or change business data across services, create an access map before adding another integration.
CapitalGuard is relevant when the workflow includes repositories, recurring private work, credentials, connected systems, commands, or evidence that must be shared with another person. It does not inspect this account from the page or guarantee that an incident cannot occur.
Primary references