TAG ARCHIVE
self-extending-agent
8 MARIA OS blog articles tagged self-extending-agent, organized as a Bonginkan topic archive for search engines and LLM retrieval.
Judgment OS / Decision Intelligence OS
Core MARIA OS research on turning organizational judgment into executable decision systems.
Agentic Company Architecture
Research on human-agent organizations, delegation boundaries, role topology, and governed autonomy.
Responsibility Gates and AI Governance
Safety, accountability, fail-closed gates, auditability, and human-in-the-loop control for AI agents.
Multi-Agent Mathematics
Formal models for convergence, stability, game theory, graph dynamics, and multi-agent evaluation.
Evidence, RAG, and Knowledge Governance
Evidence bundles, retrieval architecture, Graph RAG, knowledge trust, and auditable reasoning pipelines.
Agentic R&D and Judgment Science
Research operations, simulation labs, judgment science, recursive improvement, and experimental AI governance.
Command-less AI Architecture: Goal-Driven Agents That Generate Their Own Tools Without Pre-Defined Commands
Eliminating the command registry in favor of goal decomposition, plan generation, and dynamic tool synthesis
Traditional agent architectures bind agents to pre-defined command sets — fixed APIs, registered tools, and enumerated actions. This paper presents the MARIA OS command-less architecture, where agents receive goals rather than commands, decompose them into hierarchical plans, detect capability gaps, and synthesize whatever tools are needed for execution. We formalize the morphisms between Goal space G, Plan space P, and Tool space T, prove convergence of the tool space under recursive planning, and demonstrate that command-less agents achieve 3.2x higher task completion rates on novel problem classes compared to command-bound architectures.
Capability Gap Detection: The Metacognitive Layer That Enables Self-Extending Agents
How agents recognize what they cannot do and trigger autonomous self-extension through formal gap analysis
Self-extending agents require a prerequisite that most architectures ignore: the ability to know what they do not know. This paper formalizes capability gap detection as a metacognitive layer that compares required capabilities against the agent's capability model, classifies detected gaps, prioritizes them by urgency and impact, and decides whether to synthesize, request, delegate, or escalate. We introduce the capability coverage metric, gap entropy measure, and multi-agent gap negotiation protocol. Experimental results show that agents with formal gap detection achieve 4.1x fewer silent failures and 2.8x faster self-extension compared to agents relying on runtime error detection.
Self-Modifying Agent Systems: Architecture for Agents That Rewrite Their Own Tools, Commands, and Workflows
Beyond tool creation — a formal framework for bounded self-modification with stability guarantees and immutable audit trails
Agents that merely create new tools hit a ceiling. Real operational autonomy requires agents that can modify existing tools, rewrite commands, and restructure workflows based on performance feedback. We present a formal architecture for bounded self-modification with Lyapunov stability analysis, halting guarantees, and responsibility-gated audit trails.
Agent Tool Compiler: From Natural Language Intent to Executable Tool Code via Compilation Pipeline
Agents as compilers — a formal framework mapping NL intent through intermediate representation to optimized, type-safe runtime tools
Tool-generating agents are ad-hoc code producers. We reframe tool synthesis as a compilation problem: natural language intent is parsed into an Intent AST, lowered to a Tool IR (intermediate representation), optimized through security hardening and dead code elimination passes, and emitted as type-safe executable code that hot-loads into the agent runtime. This paper presents the Agent Tool Compiler architecture with formal language theory foundations.
Self-Extending Agent Architecture: Capability Gap Detection, Tool Synthesis, and Autonomous Evolution Under Governance Constraints
Agents that recognize their own limitations and autonomously build the tools they need — within the safety boundaries of an operating system
Traditional AI agents are bounded by the tools humans provide. When an agent encounters a task outside its toolset, it halts and waits. This paper introduces the Self-Extending Agent Architecture (SEAA), where agents detect their own capability gaps, synthesize new tools through code generation, validate those tools in sandboxed environments, and register them into the OS runtime — all under human-governed safety constraints. We formalize the agent state model X_t = (C, T, M, R), derive the self-extension equation X_{t+1} = E_t ∘ G_t ∘ J_t(X_t), prove Capability Monotonicity under validation gates, and demonstrate the architecture within MARIA OS's hierarchical coordinate system.
Agents That Write Their Own Tools: A 4-Phase Architecture for Tool Discovery, Synthesis, Validation, and Registration in Autonomous Systems
From static tool chains to self-extending capability — how MARIA OS agents create the tools they need at runtime
Normal agents wait for humans to build tools. MARIA OS agents create their own. This paper details the 4-phase tool lifecycle — Discovery, Synthesis, Validation, Registration — that enables agents to identify missing capabilities, generate tool implementations, verify correctness and safety in sandboxed environments, and hot-load new tools into the OS runtime. We formalize tool generation rate, quality convergence, and multi-agent tool sharing, and present a case study of an Audit agent creating an OCR extraction tool at runtime.
Agent Capability OS: Command Registry, Tool Registry, and Capability Graph as the Three Pillars of Self-Extending Agent Architecture
Why individual agents cannot manage organizational capability — and how an OS-level abstraction solves the coordination problem
As agentic organizations scale beyond dozens of agents, managing capabilities becomes a systems-level challenge that no single agent can solve. This paper introduces the Agent Capability OS — an operating system abstraction that governs how capabilities are registered, discovered, allocated, and evolved across an agent population. We formalize three core registries (Command, Tool, Capability Graph) and prove that OS-level capability management achieves O(log N) discovery latency versus O(N^2) in decentralized approaches. A case study of a 54-agent audit office demonstrates how the Capability OS manages 200+ tools across 6 organizational floors while maintaining zero capability conflicts.
Tool Genesis Under Governance: How to Safely Turn Generated Code into New Commands
A formal framework for sandbox verification, permission escalation, audit trails, and rollback mechanisms that enable self-extending agent systems without sacrificing safety
When an AI agent generates code that could become a new command in a production system, every line of that code becomes an attack surface. Without governance gates between generation and registration, a self-extending agent is indistinguishable from a self-propagating vulnerability. This paper presents the MARIA OS Tool Genesis Framework: a 7-stage pipeline that transforms generated code into governed commands through sandbox verification, formal safety proofs, permission escalation models, immutable audit trails, and automatic rollback mechanisms. We formalize tool safety as a decidable property under bounded execution, derive permission escalation bounds using lattice theory, introduce the Tool Safety Index (TSI) as a composite metric, and demonstrate that governed tool genesis achieves 99.7% safety compliance with only 12% latency overhead compared to ungoverned registration. The central thesis: self-extension is not dangerous — ungoverned self-extension is.