The real workflow
Where Claude enters the work
Agency teams may connect several client mailboxes, drives, knowledge sources, and project systems to a common assistant workflow.
Claude can work with conversations, files, projects, and optional connectors that retrieve from or act within services according to the user’s source-system permissions.
Documents, webpages, connector output, and MCP resources may contain instructions that conflict with the user’s goal.
Claude does not receive blanket access by default. The practical boundary is the content submitted plus the connectors, permissions, projects, and account controls the user enables.
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
chat messages, files, and project knowledge
shared chat snapshots
connectors with read or write tools
Settings to verify
Privacy and model-improvement choice
Shared chats and project visibility
Connector tool permissions and source-account scope
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, in connected workflows, the same manipulation can influence code, messages, documents, tickets, cloud actions, or data transfer across trusted systems.
Prompt injection happens when untrusted content contains instructions that compete with the user’s real request. The danger rises when the assistant can retrieve private information, call tools, run commands, or make changes.
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 account, workspace, project, device, and connected service used in this workflow.
- 2
Open suspicious material without write-capable connectors and ask Claude to identify instructions rather than follow them.
- 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
Block write tools or require approval while analyzing untrusted content. Record the result without copying private content or raw credentials into the report.
Controls to apply
Reduce access before adding trust
Block write tools or require approval while analyzing untrusted content.
Separate trusted instructions from retrieved or user-supplied content.
Use tool allowlists, denied paths, network restrictions, and approval gates around consequential actions.
Log the source of instructions and stop when tool behavior changes unexpectedly.
Decision rule
Know when a formal baseline is justified
Simple text-only use still needs judgment, but the paid security case begins when untrusted content and meaningful tool authority coexist. That is the point to map the full action-to-asset path.
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