AI agents call get_audit_trail to retrieve information from SIFTGuard without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and returns historical audit log data from the current investigation session. It performs a read-only retrieval operation without creating, modifying, deleting, or executing external operations. While audit trails may contain sensitive information about what was investigated, the tool itself has no destructive or side-effect capabilities.
From the tool's definition Tool name 'get_audit_trail' and description 'Return the complete structured audit trail of all tool calls in this investigation session' indicate data retrieval with no modification or side effects.
Attacks that exploit this kind of access
Return the complete structured audit trail of all tool calls in this investigation session. It is categorised as a Read tool in the SIFTGuard MCP Server, which means it retrieves data without modifying state.
Register the SIFTGuard MCP server in PolicyLayer and add a rule for get_audit_trail: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches SIFTGuard. Nothing to install.
get_audit_trail is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_audit_trail rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for get_audit_trail. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
get_audit_trail is provided by the SIFTGuard MCP server (sodiq-code/siftguard). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →