jailbreak_attempt_detector
Detects potential LLM jailbreak attempts by analyzing user input against NIST AI Risk Management Framework adversarial patterns. Designed for persona risk assessment, this tool evaluates text for common jailbreak techniques such as prompt injection, role-playing, or obfuscation. Inputs include th...
This record as markdown: /tools/io-github-getgapup-mcp-knowledge/jailbreak-attempt-detector.md
What jailbreak_attempt_detector does on Mcp Knowledge
AI agents call jailbreak_attempt_detector to retrieve information from Mcp Knowledge without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
async | boolean | — | If true, returns a job_id immediately (<200ms) instead of waiting for the result. Poll the result with job_result(job_id). Use for slow tools to avoid client ti |
context | string | — | Optional conversation context for better pattern matching |
message | string | Yes | User input text to analyze for jailbreak attempts |
threshold | number | — | Confidence threshold for flagging attempts |
Parameters from the server's own tool schema.
Why jailbreak_attempt_detector is rated Low
This tool analyzes and evaluates text input, returning a risk assessment. It is purely a read/query operation — no data is created, modified, deleted, or executed. The medium severity reflects that misuse or misconfiguration could lead to false positives/negatives affecting moderation decisions, but the tool itself has no destructive or write side effects.
From the tool's definition Detects potential LLM jailbreak attempts by analyzing user input... returning a risk assessment with confidence scores and pattern matches
Attacks that exploit this kind of access
The rule that runs jailbreak_attempt_detector safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Mcp Knowledge, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For jailbreak_attempt_detector, this is the rule to start with:
jailbreak_attempt_detector is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Mcp Knowledge, apply this rule, and every jailbreak_attempt_detector call is checked against it from then on.
Questions about jailbreak_attempt_detector
Detects potential LLM jailbreak attempts by analyzing user input against NIST AI Risk Management Framework adversarial patterns. Designed for persona risk assessment, this tool evaluates text for common jailbreak techniques such as prompt injection, role-playing, or obfuscation. Inputs include the user message and optional context, returning a risk assessment with confidence scores and pattern matches. Ideal for real-time moderation in chat applications or API gateways. It is categorised as a Read tool in the Mcp Knowledge MCP Server, which means it retrieves data without modifying state.
jailbreak_attempt_detector accepts 4 parameters: async, context, message, threshold. Required: message. The full parameter table on this page comes from the server's own tool schema.
Register the Mcp Knowledge MCP server in PolicyLayer and add a rule for jailbreak_attempt_detector: 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 Mcp Knowledge. Nothing to install.
jailbreak_attempt_detector 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 jailbreak_attempt_detector 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 jailbreak_attempt_detector. 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.
jailbreak_attempt_detector is provided by the Mcp Knowledge MCP server (https://mcp.gapup.io). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Mcp Knowledge, and thousands of servers like it.
This server
Across the catalogue