List all built-in injection detection patterns with categories and weights.
AI agents call injection_patterns to retrieve information from Agent Safety 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 metadata about the safety server's internal detection rules. It performs pure information retrieval with no capability to modify, execute, or delete data. The blast radius of misuse is minimal—an attacker learning the detection patterns could theoretically attempt to evade detection, but the tool itself only reads and exposes existing information without enforcing or executing anything.
From the tool's definition Tool name 'injection_patterns' combined with description 'List all built-in injection detection patterns with categories and weights' indicates a read-only operation that retrieves and enumerates existing detection patterns.
Attacks that exploit this kind of access
List all built-in injection detection patterns with categories and weights. It is categorised as a Read tool in the Agent Safety MCP Server, which means it retrieves data without modifying state.
Register the Agent Safety MCP server in PolicyLayer and add a rule for injection_patterns: 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 Agent Safety. Nothing to install.
injection_patterns 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 injection_patterns 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 injection_patterns. 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.
injection_patterns is provided by the Agent Safety MCP server (luciferforge/agent-safety-mcp). 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 →