graph TB
    subgraph Core Services
        UI[Frontend UI
React + TypeScript]
        API[Backend API
Go + GraphQL]
        DB[(Vector Store
PostgreSQL + pgvector)]
        MQ[Task Queue
Async Processing]
        Agent[AI Agents
Multi-Agent System]
    end
subgraph Knowledge Graph
        Graphiti[Graphiti
Knowledge Graph API]
        Neo4j[(Neo4j
Graph Database)]
    end
subgraph Monitoring
        Grafana[Grafana
Dashboards]
        VictoriaMetrics[VictoriaMetrics
Time-series DB]
        Jaeger[Jaeger
Distributed Tracing]
        Loki[Loki
Log Aggregation]
        OTEL[OpenTelemetry
Data Collection]
    end
subgraph Analytics
        Langfuse[Langfuse
LLM Analytics]
        ClickHouse[ClickHouse
Analytics DB]
        Redis[Redis
Cache + Rate Limiter]
        MinIO[MinIO
S3 Storage]
    end
subgraph Security Tools
        Scraper[Web Scraper
Isolated Browser]
        PenTest[Security Tools
20+ Pro Tools
Sandboxed Execution]
    end
UI --> |HTTP/WS| API
    API --> |SQL| DB
    API --> |Events| MQ
    MQ --> |Tasks| Agent
    Agent --> |Commands| PenTest
    Agent --> |Queries| DB
    Agent --> |Knowledge| Graphiti
    Graphiti --> |Graph| Neo4j
API --> |Telemetry| OTEL
    OTEL --> |Metrics| VictoriaMetrics
    OTEL --> |Traces| Jaeger
    OTEL --> |Logs| Loki
Grafana --> |Query| VictoriaMetrics
    Grafana --> |Query| Jaeger
    Grafana --> |Query| Loki
API --> |Analytics| Langfuse
    Langfuse --> |Store| ClickHouse
    Langfuse --> |Cache| Redis
    Langfuse --> |Files| MinIO
classDef core fill:#f9f,stroke:#333,stroke-width:2px,color:#000
    classDef knowledge fill:#ffa,stroke:#333,stroke-width:2px,color:#000
    classDef monitoring fill:#bbf,stroke:#333,stroke-width:2px,color:#000
    classDef analytics fill:#bfb,stroke:#333,stroke-width:2px,color:#000
    classDef tools fill:#fbb,stroke:#333,stroke-width:2px,color:#000
class UI,API,DB,MQ,Agent core
    class Graphiti,Neo4j knowledge
    class Grafana,VictoriaMetrics,Jaeger,Loki,OTEL monitoring
    class Langfuse,ClickHouse,Redis,MinIO analytics
    class Scraper,PenTest tools

erDiagram
    Flow ||--o{ Task : contains
    Task ||--o{ SubTask : contains
    SubTask ||--o{ Action : contains
    Action ||--o{ Artifact : produces
    Action ||--o{ Memory : stores
Flow {
        string id PK
        string name "Flow name"
        string description "Flow description"
        string status "active/completed/failed"
        json parameters "Flow parameters"
        timestamp created_at
        timestamp updated_at
    }
Task {
        string id PK
        string flow_id FK
        string name "Task name"
        string description "Task description"
        string status "pending/running/done/failed"
        json result "Task results"
        timestamp created_at
        timestamp updated_at
    }
SubTask {
        string id PK
        string task_id FK
        string name "Subtask name"
        string description "Subtask description"
        string status "queued/running/completed/failed"
        string agent_type "researcher/developer/executor"
        json context "Agent context"
        timestamp created_at
        timestamp updated_at
    }
Action {
        string id PK
        string subtask_id FK
        string type "command/search/analyze/etc"
        string status "success/failure"
        json parameters "Action parameters"
        json result "Action results"
        timestamp created_at
    }
Artifact {
        string id PK
        string action_id FK
        string type "file/report/log"
        string path "Storage path"
        json metadata "Additional info"
        timestamp created_at
    }
Memory {
        string id PK
        string action_id FK
        string type "observation/conclusion"
        vector embedding "Vector representation"
        text content "Memory content"
        timestamp created_at
    }

sequenceDiagram
    participant O as Orchestrator
    participant R as Researcher
    participant D as Developer
    participant E as Executor
    participant VS as Vector Store
    participant KB as Knowledge Base
Note over O,KB: Flow Initialization
    O->>VS: Query similar tasks
    VS-->>O: Return experiences
    O->>KB: Load relevant knowledge
    KB-->>O: Return context
Note over O,R: Research Phase
    O->>R: Analyze target
    R->>VS: Search similar cases
    VS-->>R: Return patterns
    R->>KB: Query vulnerabilities
    KB-->>R: Return known issues
    R->>VS: Store findings
    R-->>O: Research results
Note over O,D: Planning Phase
    O->>D: Plan attack
    D->>VS: Query exploits
    VS-->>D: Return techniques
    D->>KB: Load tools info
    KB-->>D: Return capabilities
    D-->>O: Attack plan
Note over O,E: Execution Phase
    O->>E: Execute plan
    E->>KB: Load tool guides
    KB-->>E: Return procedures
    E->>VS: Store results
    E-->>O: Execution status

graph TB
    subgraph "Long-term Memory"
        VS[(Vector Store
Embeddings DB)]
        KB[Knowledge Base
Domain Expertise]
        Tools[Tools Knowledge
Usage Patterns]
    end
subgraph "Working Memory"
        Context[Current Context
Task State]
        Goals[Active Goals
Objectives]
        State[System State
Resources]
    end
subgraph "Episodic Memory"
        Actions[Past Actions
Commands History]
        Results[Action Results
Outcomes]
        Patterns[Success Patterns
Best Practices]
    end
Context --> |Query| VS
    VS --> |Retrieve| Context
Goals --> |Consult| KB
    KB --> |Guide| Goals
State --> |Record| Actions
    Actions --> |Learn| Patterns
    Patterns --> |Store| VS
Tools --> |Inform| State
    Results --> |Update| Tools
VS --> |Enhance| KB
    KB --> |Index| VS
classDef ltm fill:#f9f,stroke:#333,stroke-width:2px,color:#000
    classDef wm fill:#bbf,stroke:#333,stroke-width:2px,color:#000
    classDef em fill:#bfb,stroke:#333,stroke-width:2px,color:#000
class VS,KB,Tools ltm
    class Context,Goals,State wm
    class Actions,Results,Patterns em

The chain summarization system manages conversation context growth by selectively summarizing older messages. This is critical for preventing token limits from being exceeded while maintaining conversation coherence.

flowchart TD
    A[Input Chain] --> B{Needs Summarization?}
    B -->|No| C[Return Original Chain]
    B -->|Yes| D[Convert to ChainAST]
    D --> E[Apply Section Summarization]
    E --> F[Process Oversized Pairs]
    F --> G[Manage Last Section Size]
    G --> H[Apply QA Summarization]
    H --> I[Rebuild Chain with Summaries]
    I --> J{Is New Chain Smaller?}
    J -->|Yes| K[Return Optimized Chain]
    J -->|No| C
classDef process fill:#bbf,stroke:#333,stroke-width:2px,color:#000
    classDef decision fill:#bfb,stroke:#333,stroke-width:2px,color:#000
    classDef output fill:#fbb,stroke:#333,stroke-width:2px,color:#000
class A,D,E,F,G,H,I process
    class B,J decision
    class C,K output

The algorithm operates on a structured representation of conversation chains (ChainAST) that preserves message types including tool calls and their responses. All summarization operations maintain critical conversation flow while reducing context size.

Global Summarizer Configuration Options

| Parameter | Environment Variable | Default | Description | | --------------------- | -------------------------------- | ------- | ---------------------------------------------------------- | | Preserve Last | SUMMARIZER_PRESERVE_LAST | true | Whether to keep all messages in the last section intact | | Use QA Pairs | SUMMARIZER_USE_QA | true | Whether to use QA pair summarization strategy | | Summarize Human in QA | SUMMARIZER_SUM_MSG_HUMAN_IN_QA | false | Whether to summarize human messages in QA pairs | | Last Section Size | SUMMARIZER_LAST_SEC_BYTES | 51200 | Maximum byte size for last section (50KB) | | Max Body Pair Size | SUMMARIZER_MAX_BP_BYTES | 16384 | Maximum byte size for a single body pair (16KB) | | Max QA Sections | SUMMARIZER_MAX_QA_SECTIONS | 10 | Maximum QA pair sections to preserve | | Max QA Size | SUMMARIZER_MAX_QA_BYTES | 65536 | Maximum byte size for QA pair sections (64KB) | | Keep QA Sections | SUMMARIZER_KEEP_QA_SECTIONS | 1 | Number of recent QA sections to keep without summarization |

Assistant Summarizer Configuration Options

Assistant instances can use customized summarization settings to fine-tune context management behavior:

| Parameter | Environment Variable | Default | Description | | ------------------ | --------------------------------------- | ------- | -------------------------------------------------------------------- | | Preserve Last | ASSISTANT_SUMMARIZER_PRESERVE_LAST | true | Whether to preserve all messages in the assistant's last section | | Last Section Size | ASSISTANT_SUMMARIZER_LAST_SEC_BYTES | 76800 | Maximum byte size for assistant's last section (75KB) | | Max Body Pair Size | ASSISTANT_SUMMARIZER_MAX_BP_BYTES | 16384 | Maximum byte size for a single body pair in assistant context (16KB) | | Max QA Sections | ASSISTANT_SUMMARIZER_MAX_QA_SECTIONS | 7 | Maximum QA sections to preserve in assistant context | | Max QA Size | ASSISTANT_SUMMARIZER_MAX_QA_BYTES | 76800 | Maximum byte size for assistant's QA sections (75KB) | | Keep QA Sections | ASSISTANT_SUMMARIZER_KEEP_QA_SECTIONS | 3 | Number of recent QA sections to preserve without summarization |

The assistant summarizer configuration provides more memory for context retention compared to the global settings, preserving more recent conversation history while still ensuring efficient token usage.

Summarizer Environment Configuration

# Default values for global summarizer logic
SUMMARIZER_PRESERVE_LAST=true
SUMMARIZER_USE_QA=true
SUMMARIZER_SUM_MSG_HUMAN_IN_QA=false
SUMMARIZER_LAST_SEC_BYTES=51200
SUMMARIZER_MAX_BP_BYTES=16384
SUMMARIZER_MAX_QA_SECTIONS=10
SUMMARIZER_MAX_QA_BYTES=65536
SUMMARIZER_KEEP_QA_SECTIONS=1
Default values for assistant summarizer logic
ASSISTANT_SUMMARIZER_PRESERVE_LAST=true
ASSISTANT_SUMMARIZER_LAST_SEC_BYTES=76800
ASSISTANT_SUMMARIZER_MAX_BP_BYTES=16384
ASSISTANT_SUMMARIZER_MAX_QA_SECTIONS=7
ASSISTANT_SUMMARIZER_MAX_QA_BYTES=76800
ASSISTANT_SUMMARIZER_KEEP_QA_SECTIONS=3

PentAGI includes sophisticated multi-layered agent supervision mechanisms to ensure efficient task execution, prevent infinite loops, and provide intelligent recovery from stuck states:

Execution Monitoring (Beta)

• Automatic Mentor Intervention: Adviser agent (mentor) is automatically invoked when execution patterns indicate potential issues
• Pattern Detection: Monitors identical tool calls (threshold: 5, configurable) and total tool calls (threshold: 10, configurable)
• Progress Analysis: Evaluates whether agent advances toward subtask objective, detects loops and inefficiencies
• Alternative Strategies: Recommends different approaches when current strategy fails
• Information Retrieval Guidance: Suggests searching for established solutions instead of reinventing
• Enhanced Response Format: Tool responses include both <original_result> and <mentor_analysis> sections
• Configurable: Enable via EXECUTION_MONITOR_ENABLED (default: false), customize thresholds with EXECUTION_MONITOR_SAME_TOOL_LIMIT and EXECUTION_MONITOR_TOTAL_TOOL_LIMIT

Performance Impact: 2-3x increase in execution time and token usage, but delivers 2x improvement in result quality based on testing with Qwen3.5-27B-FP8

Intelligent Task Planning (Beta)

• Automated Decomposition: Planner (adviser in planning mode) generates 3-7 specific, actionable steps before specialist agents begin work
• Context-Aware Plans: Analyzes full execution context via enricher agent to create informed plans
• Structured Assignment: Original request wrapped in <task_assignment> structure with execution plan and instructions
• Scope Management: Prevents scope creep by keeping agents focused on current subtask only
• Enriched Instructions: Plans highlight critical actions, potential pitfalls, and verification points
• Configurable: Enable via AGENT_PLANNING_STEP_ENABLED (default: false)

Enhanced Adviser Configuration: Works exceptionally well when adviser agent uses stronger model or enhanced settings. Example: using same base model with maximum reasoning mode for adviser (see vllm-qwen3.5-27b-fp8.provider.yml) enables comprehensive task analysis and strategic planning from identical model architecture.

Performance Impact: Adds planning overhead but significantly improves completion rates and reduces redundant work

Tool Call Limits (Always Active)

• Hard Limits: Prevent runaway executions regardless of supervision mode status
• Differentiated by Agent Type:
• General agents (Assistant, Primary Agent, Pentester, Coder, Installer): MAX_GENERAL_AGENT_TOOL_CALLS (default: 100)
• Limited agents (Searcher, Enricher, Memorist, Generator, Reporter, Adviser, Reflector, Planner): MAX_LIMITED_AGENT_TOOL_CALLS (default: 20)
• Graceful Termination: Reflector guides agents to proper completion when approaching limits
• Resource Protection: Ensures system stability and prevents resource exhaustion

Reflector Integration (Always Active)

• Automatic Correction: Invoked when LLM fails to generate tool calls after 3 attempts
• Strategic Guidance: Analyzes failures and guides agents toward proper tool usage or barrier tools (done, ask)
• Recovery Mechanism: Provides contextual guidance based on specific failure patterns
• Limit Enforcement: Coordinates graceful termination when tool call limits are reached

Recommendations for Open Source Models

**Must-Have for Models