Skip to main content

​
Sub-Agent Detection

Quint detects when AI agents spawn child agents — even without explicit instrumentation. The detection system uses three independent layers that feed signals into a correlation engine for high-confidence parent-child attribution.

​
Detection Layers

​
Layer 1: Model Divergence

When an agent uses a different model than previously seen in the same tunnel, Quint infers a sub-agent spawn. The first model seen becomes the parent model; subsequent different models trigger a split. Known divergence patterns:
Parent ModelChild ModelProvider
*opus* / *sonnet**haiku*Anthropic
*gpt-4o*gpt-4o-mini*OpenAI
*pro**flash*Google
Any model change (not just known patterns) triggers detection. Each tunnel can only split once.

​
Layer 2: Concurrency Spike

Quint tracks CONNECT tunnel counts per IP address. During the first 10 seconds (stabilization window), it learns the baseline concurrency. After that, if active tunnels exceed baseline + 2, a sub-agent is detected. 
Stabilization: 10 seconds
Spike threshold: baseline + 2 tunnels
Pending children are tracked until their model can be confirmed via Layer 1.

​
Layer 3: Temporal Gap

When time between CONNECT requests from the same IP exceeds the burst window (2000ms default), Quint evaluates:
  • Has parent trace (X-Quint-Trace header) → confirmed child (source: child_detect)
  • No trace, active tunnels exist → inferred child (source: inferred_child)
  • No trace, no active tunnels → new peer agent

​
Correlation Engine

Detection signals from all three layers merge in the correlation engine, which maintains a relationship graph with confidence scores.

​
Signal Types

SignalConstantBase Confidence
Spawn pattern matchspawn~0.85
Trace context headercontext~0.95
Temporal correlationtemporal~0.50
HMAC-verified ticketsignature1.0

​
Confidence Merging

Multiple signals for the same parent-child pair are merged using diminishing returns: 
merged = max(existing, new) + (1 - max) * 0.1
For example, a spawn signal (0.85) followed by a context signal (0.95) produces: 
max(0.85, 0.95) + (1 - 0.95) * 0.1 = 0.95 + 0.005 = 0.955

​
Relationship Graph

The engine tracks: 
type AgentRelationship struct {
    ParentAgentID string
    ChildAgentID  string
    Confidence    float64
    Signals       []Signal   // all contributing signals
    SpawnType     string     // "direct", "delegation", "fork"
    Depth         int        // nesting level
}
Relationships are published to the agent.relationships Kafka topic for downstream consumption.

​
Dashboard Visualization

The dashboard renders parent-child trees from the correlation engine data. Each node shows:
  • Agent name and provider
  • Model in use
  • Confidence score for the parent-child link
  • Signal types that contributed to detection
Sub-agent detection in forward proxy mode relies on heuristics (model divergence, timing, concurrency). For guaranteed detection, use spawn tickets with HMAC-SHA256 verification.

​
Configuration

Sub-agent detection is enabled by default in forward proxy mode. Tuning parameters:
ParameterDefaultDescription
Burst window2000msTime gap threshold for temporal detection
Stabilization window10sTime to learn baseline concurrency
Spike thresholdbaseline + 2Tunnel count increase to trigger detection
Max body preview8192 bytesHow much of POST body to read for model extraction