The real workflow
Where Claude Code enters the work
An agency coding agent can cross client boundaries when repositories, terminals, credentials, caches, or sessions are reused between engagements.
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.
Claude Code can read project files and, depending on permissions, may read beyond the working directory even when writes are narrower.
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 agencies
Agency risk compounds when staff, contractors, shared tools, and reused credentials create paths between otherwise separate client environments. In this case, for professional work, the same access can reveal contracts, pricing, unpublished plans, internal discussions, customer records, or source material covered by confidentiality obligations.
The risk is not that an AI assistant can magically see an entire device. The risk begins when a file is uploaded, a folder is granted, a project is indexed, or a connected service makes private material retrievable.
Every client remains isolated, access is attributable to a named operator, and the agency can deliver consistent evidence without revealing another client.
Context decision
Three questions before adding access
Can this operator or tool reach any repository, mailbox, drive, cache, token, or transcript belonging to another client?
Are credentials and AI sessions issued per client and person rather than shared across the agency?
Can the agency deliver useful proof to this client without including another client's names, paths, findings, or configuration?
Evidence goal: Create a separate client evidence record covering operator identity, workspace isolation, credentials, approved systems, review history, and delivery status.
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
Inspect the launch directory, additional directories, Read deny rules, sandbox boundaries, and trusted-directory state.
- 3
Assign the decision and next review to the client service owner or agency security lead; do not leave the access boundary as an unwritten assumption.
- 4
Start inside a dedicated project and deny home, credential, client, and unrelated workspace paths. Record the result without copying private content or raw credentials into the report.
Controls to apply
Reduce access before adding trust
Start inside a dedicated project and deny home, credential, client, and unrelated workspace paths.
Separate sensitive work from ordinary AI-ready material before granting access.
Prefer the smallest folder, file, or project scope that completes the task.
Remove stale uploads and connections, then document who should review access again and when.
Decision rule
Know when a formal baseline is justified
If the tool only receives public or disposable material, use the free checklist. If it can reach recurring private work, repositories, or client files, create a documented access baseline before the next sensitive task.
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