What can AI read?
Find sensitive configs, prompts, logs, and files that may enter agent context.
CapitalGuard features
Know what your AI tools can reach.
AI coding tools move faster with access. CapitalGuard shows where that access can expose your business before the next prompt, plugin, or automation expands it.
CapitalGuard AccessGraph
Read. Change. Execute. Transfer.
Repository workspaceIllustrative coding-agent policy
ReadCredential-bearing pathsBlock
ChangeAuth and billing logicHuman approval
ExecuteLocal commandsHuman approval
TransferPrompts and logsRedact
Licensed scans generate this map locally from aggregate exposure state without placing secret values in the graph.CGX-8D42-7AC1-55F0
Three questions that matter
Answer them before an incident does.
What can AI change?
Map workflows, protected paths, and high-impact edits that need human approval.
What can leave the repo?
Spot commands and tool bridges that can move sensitive context outside your control.
From uncertainty to a clear next move.
AccessGraph
See the path from AI tool to sensitive asset, risky action, and required control.
Secret-risk scan
Find secret-like exposure without printing the sensitive value.
Instruction-risk detection
Flag repository text that could quietly steer an agent toward unsafe behavior.
Installable guardrails
Create protected paths, redaction rules, command limits, and approval gates.
Guardprint changes
Know when the repo's aggregate access state changes after the baseline.
Decision-ready report
Give founders and teams the risk, business impact, and next precaution in plain language.
Before and after
Stop guessing. Start controlling access.
Before
AI tools have broad access nobody has mapped.
With CapitalGuard
Read, change, execute, and transfer paths are visible.
Before
Risk is buried in files, prompts, and workflows.
With CapitalGuard
Evidence is redacted, ranked, and tied to business impact.
Before
The team knows something feels unsafe.
With CapitalGuard
The team knows which guardrail should come next.
Choose your scope
One repo, a growing team, or every client.
1 repo scan
Starter License
Planned regular €199Save €50
€149
Protect everyday AI use and your first repository.
1 repo scanCapitalGuard Risk ScoreAccessGraph snapshot and top findings
View Starter5 repo scans
Pro License
Planned regular €699Save €200
€499
Make AI-agent exposure visible, governable, and provable.
5 repo scansAccessGraph change history and license consoleVersioned A-BOM and event-chain evidence
View Pro25 repo scans
Agency License
Planned regular €2,799Save €800
€1,999
Turn AI-agent security into a client-ready delivery system.
25 repo scansClient AccessGraph portfolioClient A-BOM and tamper-evident scan evidence
View Agency