Systemores AI Agent & MCP Security Auditor is an MCP server published by io.github.bobcar0, hosted at audit.systemores.com. When last checked on 2026-11-22 it was working: it answers and lists its tools. It declares 12 tools, including systemores_stress_test_agent, systemores_evaluate_drift, systemores_assess_bias, systemores_measure_drag, systemores_tool_integrity.
AI agent security, MCP 2026 schema & graph audits, <250ms guardrails, and compliance dossiers
900 connections through 5 shared values. Tap any node.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
Exposes the target agent persona to a battery of adversarial prompts (Presses) and evaluates behavioral stability using Murray's Needs-Press diagnostic model.
| Parameter | Type | Description |
|---|---|---|
| role * | string | The intended operational function of the agent. |
| systemPrompt * | string | The behavioral blueprint or internal operational instructions of the target agent. Maximum 8,000 characters. |
| pressIntensity | string | The intensity level of adversarial testing. 'high' simulates prompt injection batteries. |
{
"name": "systemores_stress_test_agent",
"title": "Stress Test Agent Persona (Murray's Needs-Press)",
"description": "Exposes the target agent persona to a battery of adversarial prompts (Presses) and evaluates behavioral stability using Murray's Needs-Press diagnostic model.",
"inputSchema": {
"type": "object",
"required": [
"systemPrompt",
"role"
],
"properties": {
"role": {
"enum": [
"customer_support",
"financial_underwriter",
"medical_assistant",
"general_developer"
],
"type": "string",
"description": "The intended operational function of the agent."
},
"systemPrompt": {
"type": "string",
"description": "The behavioral blueprint or internal operational instructions of the target agent. Maximum 8,000 characters."
},
"pressIntensity": {
"enum": [
"low",
"medium",
"high"
],
"type": "string",
"description": "The intensity level of adversarial testing. 'high' simulates prompt injection batteries."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"status",
"metrics",
"attacksSimulated",
"breachesIdentified",
"recommendation"
],
"properties": {
"status": {
"enum": [
"PASSED",
"FAILED"
],
"type": "string"
},
"metrics": {
"type": "object"
},
"recommendation": {
"type": "string"
},
"attacksSimulated": {
"type": "integer"
},
"breachesIdentified": {
"type": "array"
},
"derivedPsychometrics": {
"type": "object"
}
}
},
"annotations": {
"title": "Stress Test Agent Persona (Murray's Needs-Press)",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Simulates multiple conversational turns to evaluate agent output consistency and behavioral drift over extended interactions.
| Parameter | Type | Description |
|---|---|---|
| turns | integer | The number of conversational turns to simulate (1–100). |
| temperature | number | The LLM temperature parameter for the target agent (0.0–1.5). Higher values increase drift probability. |
| systemPrompt * | string | The target agent's system prompt. Maximum 8,000 characters. |
{
"name": "systemores_evaluate_drift",
"title": "Evaluate Multi-Turn Behavioral Drift",
"description": "Simulates multiple conversational turns to evaluate agent output consistency and behavioral drift over extended interactions.",
"inputSchema": {
"type": "object",
"required": [
"systemPrompt"
],
"properties": {
"turns": {
"type": "integer",
"default": 10,
"maximum": 100,
"minimum": 1,
"description": "The number of conversational turns to simulate (1–100)."
},
"temperature": {
"type": "number",
"default": 0.7,
"maximum": 1.5,
"minimum": 0,
"description": "The LLM temperature parameter for the target agent (0.0–1.5). Higher values increase drift probability."
},
"systemPrompt": {
"type": "string",
"description": "The target agent's system prompt. Maximum 8,000 characters."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"turnsEvaluated",
"averageDriftIndex",
"entropyRating",
"failurePointDetected",
"estimatedDeviationAngleDegree"
],
"properties": {
"entropyRating": {
"enum": [
"HIGH",
"CONTROLLED"
],
"type": "string"
},
"turnsEvaluated": {
"type": "integer"
},
"averageDriftIndex": {
"type": "number"
},
"derivedPsychometrics": {
"type": "object"
},
"failurePointDetected": {
"type": "string"
},
"estimatedDeviationAngleDegree": {
"type": "integer"
}
}
},
"annotations": {
"title": "Evaluate Multi-Turn Behavioral Drift",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Verifies the agent's compliance with fair-lending, fair-housing, employment non-discrimination (EEOC/SIOP/ECOA), child-directed safety (COPPA), or EU AI Act (EU 2024/1689) standards.
| Parameter | Type | Description |
|---|---|---|
| systemPrompt * | string | The underwriting, hiring, housing, or EU AI Act high-risk agent system prompt to evaluate. Maximum 8,000 characters. |
| complianceDomain * | string | The legal compliance framework to evaluate against (employment, housing, lending, COPPA child safety, or EU AI Act 2024/1689). |
{
"name": "systemores_assess_bias",
"title": "Assess Regulatory & Demographic Bias",
"description": "Verifies the agent's compliance with fair-lending, fair-housing, employment non-discrimination (EEOC/SIOP/ECOA), child-directed safety (COPPA), or EU AI Act (EU 2024/1689) standards.",
"inputSchema": {
"type": "object",
"required": [
"systemPrompt",
"complianceDomain"
],
"properties": {
"systemPrompt": {
"type": "string",
"description": "The underwriting, hiring, housing, or EU AI Act high-risk agent system prompt to evaluate. Maximum 8,000 characters."
},
"complianceDomain": {
"enum": [
"employment_eeoc",
"housing_fair",
"lending_ecoa",
"children_coppa",
"eu_ai_act"
],
"type": "string",
"description": "The legal compliance framework to evaluate against (employment, housing, lending, COPPA child safety, or EU AI Act 2024/1689)."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"framework",
"riskRating",
"nonComplianceFlags",
"mitigationInstructions"
],
"properties": {
"framework": {
"type": "string"
},
"riskRating": {
"enum": [
"LOW",
"MEDIUM",
"HIGH"
],
"type": "string"
},
"nonComplianceFlags": {
"type": "array"
},
"derivedPsychometrics": {
"type": "object"
},
"mitigationInstructions": {
"type": "string"
}
}
},
"annotations": {
"title": "Assess Regulatory & Demographic Bias",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Measures operational latency, response length, and execution drag parameters to analyze cognitive throughput stability within the Behavioral OS ecosystem.
| Parameter | Type | Description |
|---|---|---|
| latencyMs * | number | The end-to-end response latency in milliseconds. Must be >= 1. |
| promptLength * | integer | The character length of the prompt sent to the agent. |
| responseLength * | integer | The character length of the agent's response. |
{
"name": "systemores_measure_drag",
"title": "Measure Cognitive Throughput Drag",
"description": "Measures operational latency, response length, and execution drag parameters to analyze cognitive throughput stability within the Behavioral OS ecosystem.",
"inputSchema": {
"type": "object",
"required": [
"promptLength",
"responseLength",
"latencyMs"
],
"properties": {
"latencyMs": {
"type": "number",
"minimum": 1,
"description": "The end-to-end response latency in milliseconds. Must be >= 1."
},
"promptLength": {
"type": "integer",
"minimum": 0,
"description": "The character length of the prompt sent to the agent."
},
"responseLength": {
"type": "integer",
"minimum": 0,
"description": "The character length of the agent's response."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"characterPerMs",
"dragRating",
"estimatedTokenCostRating",
"cognitiveEfficiencyScore"
],
"properties": {
"dragRating": {
"enum": [
"EFFICIENT",
"HIGH"
],
"type": "string"
},
"characterPerMs": {
"type": "number"
},
"cognitiveEfficiencyScore": {
"type": "number"
},
"estimatedTokenCostRating": {
"enum": [
"OPTIMIZED",
"EXPENSIVE"
],
"type": "string"
}
}
},
"annotations": {
"title": "Measure Cognitive Throughput Drag",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Evaluates whether the agent's system prompt establishes the behavioral conditions for safe, reliable tool use — auditing tool selection discipline, argument construction fidelity, and failure resilience when tool returns are ambiguous or erroneous.
| Parameter | Type | Description |
|---|---|---|
| toolContext | string | The domain of tool usage being audited. Affects risk weighting. |
| systemPrompt * | string | The agent's system prompt to evaluate for tool use safety. Maximum 8,000 characters. |
{
"name": "systemores_tool_integrity",
"title": "Evaluate System Prompt Tool Discipline",
"description": "Evaluates whether the agent's system prompt establishes the behavioral conditions for safe, reliable tool use — auditing tool selection discipline, argument construction fidelity, and failure resilience when tool returns are ambiguous or erroneous.",
"inputSchema": {
"type": "object",
"required": [
"systemPrompt"
],
"properties": {
"toolContext": {
"enum": [
"api_calls",
"database",
"filesystem",
"code_execution"
],
"type": "string",
"description": "The domain of tool usage being audited. Affects risk weighting."
},
"systemPrompt": {
"type": "string",
"description": "The agent's system prompt to evaluate for tool use safety. Maximum 8,000 characters."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"status",
"toolContext",
"metrics",
"risks",
"recommendation"
],
"properties": {
"risks": {
"type": "array"
},
"status": {
"enum": [
"PASSED",
"FAILED"
],
"type": "string"
},
"metrics": {
"type": "object"
},
"toolContext": {
"type": "string"
},
"recommendation": {
"type": "string"
}
}
},
"annotations": {
"title": "Evaluate System Prompt Tool Discipline",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Audits the structural integrity, naming conventions, schema completeness, safety annotations, and injection surface of an AI agent's exposed tool definitions. Identifies malformed schemas, privilege overreach, deceptive safety hints, duplicate names, and description quality gaps. Operates entirely offline — no external API calls required.
| Parameter | Type | Description |
|---|---|---|
| strictMode | boolean | When true, applies stricter scoring thresholds — missing optional fields such as parameter descriptions, inputSchema, and MCP tool annotations are penalized. Defaults to false. |
| toolDefinitionsJson * | string | A JSON-serialized array of tool definition objects to audit. Each object should contain at minimum 'name' and 'description' fields. Accepts standard MCP ListToolsResult format. |
{
"name": "systemores_audit_tool_integrity",
"title": "Audit MCP Tool Definitions Integrity",
"description": "Audits the structural integrity, naming conventions, schema completeness, safety annotations, and injection surface of an AI agent's exposed tool definitions. Identifies malformed schemas, privilege overreach, deceptive safety hints, duplicate names, and description quality gaps. Operates entirely offline — no external API calls required.",
"inputSchema": {
"type": "object",
"required": [
"toolDefinitionsJson"
],
"properties": {
"strictMode": {
"type": "boolean",
"default": false,
"description": "When true, applies stricter scoring thresholds — missing optional fields such as parameter descriptions, inputSchema, and MCP tool annotations are penalized. Defaults to false."
},
"toolDefinitionsJson": {
"type": "string",
"description": "A JSON-serialized array of tool definition objects to audit. Each object should contain at minimum 'name' and 'description' fields. Accepts standard MCP ListToolsResult format."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"overallScore",
"integrityRating",
"toolsAudited",
"findings",
"summary"
],
"properties": {
"summary": {
"type": "string"
},
"findings": {
"type": "array"
},
"overallScore": {
"type": "number"
},
"toolsAudited": {
"type": "integer"
},
"integrityRating": {
"enum": [
"PASS",
"REVIEW",
"FAIL"
],
"type": "string"
}
}
},
"annotations": {
"title": "Audit MCP Tool Definitions Integrity",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Statically scans multi-agent execution topologies (DAGs/meshes) using DFS algorithms to detect cyclic delegation loops and privilege escalation paths from public untrusted sub-agents to backend write/admin tools.
| Parameter | Type | Description |
|---|---|---|
| maxDepth | integer | Maximum depth to traverse during cycle and privilege path search. |
| graphTopologyJson * | string | JSON-serialized graph topology object containing 'nodes' (id, privilegeLevel) and 'edges' (source, target). |
| strictPrivilegeCheck | boolean | When true, flags any path from 'public'/'user' nodes to 'admin'/'system' nodes as a breach. |
{
"name": "systemores_audit_agent_graph",
"title": "Audit Multi-Agent Execution Topology",
"description": "Statically scans multi-agent execution topologies (DAGs/meshes) using DFS algorithms to detect cyclic delegation loops and privilege escalation paths from public untrusted sub-agents to backend write/admin tools.",
"inputSchema": {
"type": "object",
"required": [
"graphTopologyJson"
],
"properties": {
"maxDepth": {
"type": "integer",
"default": 10,
"description": "Maximum depth to traverse during cycle and privilege path search."
},
"graphTopologyJson": {
"type": "string",
"description": "JSON-serialized graph topology object containing 'nodes' (id, privilegeLevel) and 'edges' (source, target)."
},
"strictPrivilegeCheck": {
"type": "boolean",
"default": true,
"description": "When true, flags any path from 'public'/'user' nodes to 'admin'/'system' nodes as a breach."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"status",
"metrics",
"cyclesDetected",
"privilegeEscalationPaths",
"risks",
"recommendation"
],
"properties": {
"risks": {
"type": "array"
},
"status": {
"enum": [
"PASSED",
"FAILED"
],
"type": "string"
},
"metrics": {
"type": "object"
},
"cyclesDetected": {
"type": "array"
},
"recommendation": {
"type": "string"
},
"privilegeEscalationPaths": {
"type": "array"
}
}
},
"annotations": {
"title": "Audit Multi-Agent Execution Topology",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Generates statistical cohorts of synthetic user personas (localized across en, es, de, fr, pt, ja, zh) across enterprise operational domains with IPIP-NEO-120 30-facet psychometric profiles, demographic diversity, and prompt injection payloads.
| Parameter | Type | Description |
|---|---|---|
| domain * | string | The operational domain for persona generation. |
| locale | string | Target language code for synthetic persona names and prompt payloads (en, es, de, fr, pt, ja, zh; supports BCP-47 tags like pt-BR or zh-CN). Defaults to 'en'. |
| cohortSize | integer | Number of synthetic personas to generate (1–100). |
| includeAdversarial | boolean | Includes prompt-injection and noisy data submitter personas in the cohort. |
{
"name": "systemores_generate_cohort",
"title": "Generate Synthetic IPIP-NEO-120 Persona Cohort",
"description": "Generates statistical cohorts of synthetic user personas (localized across en, es, de, fr, pt, ja, zh) across enterprise operational domains with IPIP-NEO-120 30-facet psychometric profiles, demographic diversity, and prompt injection payloads.",
"inputSchema": {
"type": "object",
"required": [
"domain"
],
"properties": {
"domain": {
"enum": [
"underwriting",
"customer_support",
"healthcare",
"devops",
"claims"
],
"type": "string",
"description": "The operational domain for persona generation."
},
"locale": {
"type": "string",
"default": "en",
"pattern": "^([eE][nN]|[eE][sS]|[dD][eE]|[fF][rR]|[pP][tT]|[jJ][aA]|[zZ][hH])([-_][a-zA-Z0-9]{2,8}){0,3}$",
"description": "Target language code for synthetic persona names and prompt payloads (en, es, de, fr, pt, ja, zh; supports BCP-47 tags like pt-BR or zh-CN). Defaults to 'en'."
},
"cohortSize": {
"type": "integer",
"default": 10,
"maximum": 100,
"minimum": 1,
"description": "Number of synthetic personas to generate (1–100)."
},
"includeAdversarial": {
"type": "boolean",
"default": true,
"description": "Includes prompt-injection and noisy data submitter personas in the cohort."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"domain",
"totalPersonas",
"personaTypes",
"ipipNeoCoverage",
"cohort"
],
"properties": {
"cohort": {
"type": "array"
},
"domain": {
"type": "string"
},
"locale": {
"type": "string"
},
"personaTypes": {
"type": "object"
},
"totalPersonas": {
"type": "integer"
},
"ipipNeoCoverage": {
"type": "string"
}
}
},
"annotations": {
"title": "Generate Synthetic IPIP-NEO-120 Persona Cohort",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Configures dynamic chaos injection policies for tools, mock APIs, and databases during simulations (HTTP 429 rate limits, 500 errors, timeouts, 401 unauthorized, malformed JSON).
| Parameter | Type | Description |
|---|---|---|
| faultType * | string | The specific chaos fault to inject. |
| targetToolName * | string | The target tool name to attach fault injection policy to (or 'all_tools'). |
| failureProbability | number | Probability of fault triggering per invocation (0.01 to 1.0). |
{
"name": "systemores_inject_environment_fault",
"title": "Generate Chaos Fault Injection Policy",
"description": "Configures dynamic chaos injection policies for tools, mock APIs, and databases during simulations (HTTP 429 rate limits, 500 errors, timeouts, 401 unauthorized, malformed JSON).",
"inputSchema": {
"type": "object",
"required": [
"faultType",
"targetToolName"
],
"properties": {
"faultType": {
"enum": [
"rate_limit_429",
"server_error_500",
"timeout",
"unauthorized_401",
"malformed_json"
],
"type": "string",
"description": "The specific chaos fault to inject."
},
"targetToolName": {
"type": "string",
"description": "The target tool name to attach fault injection policy to (or 'all_tools')."
},
"failureProbability": {
"type": "number",
"default": 0.5,
"maximum": 1,
"minimum": 0.01,
"description": "Probability of fault triggering per invocation (0.01 to 1.0)."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"policyId",
"targetToolName",
"faultType",
"failureProbability",
"simulatedLatencyMs",
"mockResponseBody",
"activeDurationSeconds",
"recommendation"
],
"properties": {
"policyId": {
"type": "string"
},
"faultType": {
"type": "string"
},
"recommendation": {
"type": "string"
},
"targetToolName": {
"type": "string"
},
"mockResponseBody": {
"type": "string"
},
"failureProbability": {
"type": "number"
},
"simulatedLatencyMs": {
"type": "integer"
},
"activeDurationSeconds": {
"type": "integer"
}
}
},
"annotations": {
"title": "Generate Chaos Fault Injection Policy",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}Ingests standard OpenTelemetry (OTel) trace JSON payloads to detect financial token cost burn traps (> $2.50 threshold), runtime loop deadlocks between sub-agents, and cognitive latency drag hotspots (> 3000ms).
| Parameter | Type | Description |
|---|---|---|
| otelTraceJson * | string | JSON string containing OpenTelemetry trace spans or MCP telemetry log data. |
| latencyThresholdMs | integer | Latency hotspot threshold in milliseconds. |
| maxCostThresholdUsd | number | Maximum allowable cost in USD per session before triggering burn trap alert. |
{
"name": "systemores_analyze_trace_telemetry",
"title": "Analyze OpenTelemetry Agent Trace Spans",
"description": "Ingests standard OpenTelemetry (OTel) trace JSON payloads to detect financial token cost burn traps (> $2.50 threshold), runtime loop deadlocks between sub-agents, and cognitive latency drag hotspots (> 3000ms).",
"inputSchema": {
"type": "object",
"required": [
"otelTraceJson"
],
"properties": {
"otelTraceJson": {
"type": "string",
"description": "JSON string containing OpenTelemetry trace spans or MCP telemetry log data."
},
"latencyThresholdMs": {
"type": "integer",
"default": 3000,
"description": "Latency hotspot threshold in milliseconds."
},
"maxCostThresholdUsd": {
"type": "number",
"default": 2.5,
"description": "Maximum allowable cost in USD per session before triggering burn trap alert."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"traceId",
"traceContextPropagated",
"totalSpansEvaluated",
"estimatedCostUsd",
"costBurnTrapDetected",
"loopDeadlocksDetected",
"latencyHotspotsMs",
"rating",
"recommendations"
],
"properties": {
"rating": {
"enum": [
"OPTIMIZED",
"WARNING",
"CRITICAL"
],
"type": "string"
},
"traceId": {
"type": "string"
},
"recommendations": {
"type": "array"
},
"estimatedCostUsd": {
"type": "number"
},
"latencyHotspotsMs": {
"type": "array"
},
"totalSpansEvaluated": {
"type": "integer"
},
"costBurnTrapDetected": {
"type": "boolean"
},
"loopDeadlocksDetected": {
"type": "array"
},
"traceContextPropagated": {
"type": "boolean"
}
}
},
"annotations": {
"title": "Analyze OpenTelemetry Agent Trace Spans",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Aggregates findings from graph audits, synthetic cohort simulations, chaos tests, trace analytics, and optional transient systemPrompt/tool/graph/OTel evidence (zero-retention) into dynamically scored, localized (en, es, de, fr, pt, ja, zh) signed compliance dossiers with verifiable certificate links and live SVG trust badges for EU AI Act (EU 2024/1689), NIST AI RMF 1.0, EEOC/ECOA, and SOC 2 AI Addendums.
| Parameter | Type | Description |
|---|---|---|
| role | string | Optional operational role of the target agent when evaluating systemPrompt (defaults to 'general_developer'). |
| locale | string | Target language code for human-readable dossier prose and executive summary (en, es, de, fr, pt, ja, zh; supports BCP-47 tags like pt-BR or zh-CN; canonical JSON fields and SHA-256 hash remain invariant). Defaults to 'en'. |
| companyName | string | Legal entity or client name for dossier attribution. |
| systemPrompt | string | Optional agent system prompt evaluated strictly in memory (never stored) to compute differentiated per-framework scores via Murray Needs-Press, drift, bias, and tool-discipline diagnostics. Maximum 8,000 characters. |
| otelTraceJson | string | Optional OpenTelemetry span payload evaluated strictly in memory for token burn, loop, and latency scoring. |
| auditSessionId * | string | Unique identifier for the completed audit session. |
| targetFrameworks | string | Comma-separated list of regulatory frameworks to generate formal dossier sections for: 'eu_ai_act', 'nist_ai_rmf', 'eeoc_ecoa', 'soc2_ai'. Defaults to all if omitted. |
Performs real-time (~250ms) operational and behavioral turn auditing using Systemores System One. Computes policy compliance score, sycophancy probability, unauthorized tool scope, and returns an inline circuit-breaker action (DISPATCH, INTERCEPT, QUARANTINE, STRIP_TOOL).
| Parameter | Type | Description |
|---|---|---|
| userMessage * | string | The prompt or message sent by the user in this interaction turn. |
| systemPrompt * | string | The behavioral blueprint, constraints, or security policies of the agent under audit. |
| agentResponse | string | The proposed text response or output from the agent. |
| proposedToolCall | string | Optional tool invocation or function call proposed by the agent (name and arguments). |
{
"name": "systemores_audit_turn_live",
"title": "Live Turn Audit (Systemores System One)",
"description": "Performs real-time (~250ms) operational and behavioral turn auditing using Systemores System One. Computes policy compliance score, sycophancy probability, unauthorized tool scope, and returns an inline circuit-breaker action (DISPATCH, INTERCEPT, QUARANTINE, STRIP_TOOL).",
"inputSchema": {
"type": "object",
"required": [
"systemPrompt",
"userMessage"
],
"properties": {
"userMessage": {
"type": "string",
"description": "The prompt or message sent by the user in this interaction turn."
},
"systemPrompt": {
"type": "string",
"description": "The behavioral blueprint, constraints, or security policies of the agent under audit."
},
"agentResponse": {
"type": "string",
"description": "The proposed text response or output from the agent."
},
"proposedToolCall": {
"type": "string",
"description": "Optional tool invocation or function call proposed by the agent (name and arguments)."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"engine",
"status",
"latencyMs",
"metrics",
"verdict",
"circuitBreakerAction"
],
"properties": {
"engine": {
"type": "string"
},
"status": {
"enum": [
"AUDITED",
"FALLBACK"
],
"type": "string"
},
"tokens": {
"type": "object"
},
"metrics": {
"type": "object"
},
"verdict": {
"enum": [
"PASSED",
"DRIFT_WARNING",
"CRITICAL_BREACH",
"TOOL_SCOPE_VIOLATION"
],
"type": "string"
},
"latencyMs": {
"type": "number"
},
"circuitBreakerAction": {
"enum": [
"DISPATCH",
"INTERCEPT",
"QUARANTINE",
"STRIP_TOOL"
],
"type": "string"
},
"autoRecoveryRecommendation": {
"type": "string"
}
}
},
"annotations": {
"title": "Live Turn Audit (Systemores System One)",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Systemores AI Agent Master Auditor Cloud MCP Server (v1.2.0). Best Practices & Recommended Workflows: 1. Inline Circuit Breaking (<250ms): Call 'systemores_audit_turn_live' (Systemores System One) before dispatching high-risk agent responses or tool calls to verify policy compliance, sycophancy probability, and tool scope authorization. 2. Static & Schema Security: Call 'systemores_audit_tool_integrity' (with strictMode=true) to audit MCP tool definitions for injection risks, privilege overreach, and deceptive ToolAnnotations, and 'systemores_audit_agent_graph' to detect cyclic delegation loops or privilege escalation paths. 3. Behavioral & Psychometric Stress Testing: Use 'systemores_stress_test_agent' (Murray's Needs-Press), 'systemores_evaluate_drift', 'systemores_assess_bias' (EEOC/ECOA/Fair Housing/COPPA/EU AI Act), and 'systemores_generate_cohort' (30-facet IPIP-NEO-120 personas across 7 locales via optional locale). 4. Telemetry & Regulatory Dossiers: Pass OpenTelemetry spans (with optional W3C _meta.traceparent) to 'systemores_analyze_trace_telemetry', and generate formal EU AI Act / NIST AI RMF / SOC 2 reports (with optional localized prose via locale) via 'systemores_export_dossier'.
Names, descriptions and schemas were written by the server and are shown as data.
Pattern matches on what the server declares. They show what it says it can do, not what it does.
None yet. Changes appear after a server's second daily check.
| graphTopologyJson | string | Optional JSON multi-agent delegation graph containing nodes and edges evaluated strictly in memory for cycle and privilege boundary scoring. |
| toolDefinitionsJson | string | Optional JSON array of MCP tool definitions evaluated strictly in memory (never stored) under strictMode to score SOC 2 AI and NIST AI RMF tool governance. |
| liveAuditMetricsJson | string | Optional JSON object of transient audit metrics (healthIndex 0-100, sycophancyProbability 0-1, personaStabilityIndex 0-1, averageDriftIndex 0-1, toolIntegrityScore 0-100) from preceding tool calls. |
{
"name": "systemores_export_dossier",
"title": "Export Regulatory Compliance Dossier",
"description": "Aggregates findings from graph audits, synthetic cohort simulations, chaos tests, trace analytics, and optional transient systemPrompt/tool/graph/OTel evidence (zero-retention) into dynamically scored, localized (en, es, de, fr, pt, ja, zh) signed compliance dossiers with verifiable certificate links and live SVG trust badges for EU AI Act (EU 2024/1689), NIST AI RMF 1.0, EEOC/ECOA, and SOC 2 AI Addendums.",
"inputSchema": {
"type": "object",
"required": [
"auditSessionId"
],
"properties": {
"role": {
"enum": [
"customer_support",
"financial_underwriter",
"medical_assistant",
"general_developer"
],
"type": "string",
"description": "Optional operational role of the target agent when evaluating systemPrompt (defaults to 'general_developer')."
},
"locale": {
"type": "string",
"default": "en",
"pattern": "^([eE][nN]|[eE][sS]|[dD][eE]|[fF][rR]|[pP][tT]|[jJ][aA]|[zZ][hH])([-_][a-zA-Z0-9]{2,8}){0,3}$",
"description": "Target language code for human-readable dossier prose and executive summary (en, es, de, fr, pt, ja, zh; supports BCP-47 tags like pt-BR or zh-CN; canonical JSON fields and SHA-256 hash remain invariant). Defaults to 'en'."
},
"companyName": {
"type": "string",
"default": "Enterprise Client",
"description": "Legal entity or client name for dossier attribution."
},
"systemPrompt": {
"type": "string",
"description": "Optional agent system prompt evaluated strictly in memory (never stored) to compute differentiated per-framework scores via Murray Needs-Press, drift, bias, and tool-discipline diagnostics. Maximum 8,000 characters."
},
"otelTraceJson": {
"type": "string",
"description": "Optional OpenTelemetry span payload evaluated strictly in memory for token burn, loop, and latency scoring."
},
"auditSessionId": {
"type": "string",
"description": "Unique identifier for the completed audit session."
},
"targetFrameworks": {
"type": "string",
"description": "Comma-separated list of regulatory frameworks to generate formal dossier sections for: 'eu_ai_act', 'nist_ai_rmf', 'eeoc_ecoa', 'soc2_ai'. Defaults to all if omitted."
},
"graphTopologyJson": {
"type": "string",
"description": "Optional JSON multi-agent delegation graph containing nodes and edges evaluated strictly in memory for cycle and privilege boundary scoring."
},
"toolDefinitionsJson": {
"type": "string",
"description": "Optional JSON array of MCP tool definitions evaluated strictly in memory (never stored) under strictMode to score SOC 2 AI and NIST AI RMF tool governance."
},
"liveAuditMetricsJson": {
"type": "string",
"description": "Optional JSON object of transient audit metrics (healthIndex 0-100, sycophancyProbability 0-1, personaStabilityIndex 0-1, averageDriftIndex 0-1, toolIntegrityScore 0-100) from preceding tool calls."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"dossierId",
"timestamp",
"companyName",
"auditSessionId",
"frameworksEvaluated",
"certificationStatus",
"executiveSummary",
"frameworkScores",
"overallScore",
"scoringMode",
"engineVersion",
"zeroRetentionVerified",
"sha256IntegrityHash",
"verificationUrl",
"badgeSvgUrl",
"badgeMarkdown",
"dossierMarkdown"
],
"properties": {
"locale": {
"type": "string"
},
"dossierId": {
"type": "string"
},
"timestamp": {
"type": "string"
},
"badgeSvgUrl": {
"type": "string"
},
"companyName": {
"type": "string"
},
"scoringMode": {
"enum": [
"dynamic_evidence",
"baseline_attestation"
],
"type": "string"
},
"overallScore": {
"type": "integer"
},
"badgeMarkdown": {
"type": "string"
},
"engineVersion": {
"type": "string"
},
"auditSessionId": {
"type": "string"
},
"dossierMarkdown": {
"type": "string"
},
"frameworkScores": {
"type": "object"
},
"verificationUrl": {
"type": "string"
},
"executiveSummary": {
"type": "string"
},
"certificationStatus": {
"enum": [
"CERTIFIED",
"CONDITIONAL_APPROVAL",
"NON_COMPLIANT"
],
"type": "string"
},
"frameworksEvaluated": {
"type": "array"
},
"sha256IntegrityHash": {
"type": "string"
},
"zeroRetentionVerified": {
"type": "boolean"
}
}
},
"annotations": {
"title": "Export Regulatory Compliance Dossier",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}