Memory · Software component

Prompt Context Builder

Software componentMemoryCognition & Memoryarc:PromptContextBuilder

A memory component that rebuilds the complete prompt context (goal, completed-step results, current state, requested action) from external state before every stateless LLM invocation.

Responsibility. Reconstructs full LLM prompt context from agent state for each call.

Also known as: Context reconstruction, Message history reconstruction, Context assembly

Direct neighbourhood (hover for relationship types)

Relationships

is configured by structural

invokes dependency

is invoked by dependency

reads dependency

Conversation State Store abstract Ch3.10 Ch5.9
Episodic Memory Store Ch1.6
Prompt Exemplar Set Ch5.9 Ch8.3
System Prompt Template Ch5.9 Ch8.3
Tool Schema Ch8.3
Working Memory Buffer abstract Ch1.6

writes dependency

Working Memory Buffer abstract Ch5.9

receives data from dynamic

sends data to dynamic

LLM Inference Service Ch3.10

is constrained by control

Context Budget Allocator Ch5.9
Context Window Manager abstract Ch3.10

Design guidance

MUST supply the LLM with full context on every invocation, reconstructed from external state, because each inference request is stateless.
SHOULD combine detailed short-term context with relevant long-term summaries when building the prompt.
SHOULD NOT fill the context window completely during assembly; reserve capacity for intermediate reasoning and generated output.
SHOULD retrieve the most information-dense sources rather than the longest or most comprehensive documents.
SHOULD assemble prompts following a static-first layout so provider prefix caching applies.
SHOULD pull reference data through query-driven retrieval instead of injecting whole static datasets into every request.

Quantitative guidance

As stated by the sources; verify before use.

Research-agent example (128k window): 800 system + 2,500 history + 60,000 retrieved + 50 query = 63,350 tokens (49.5%) after assembly; 64,650 tokens left for reasoning and output (Ch5.9).

Classification

Patterns: Explicit context reconstructionInjecting retrieved episode summaries into the promptAssembly order: system prompt, conversation history, retrieved context, user inputReserve budget for reasoning traces and output during assembly
Technologies: LangChain AgentExecutorLangGraphAutoGen
Quality attributes: Functional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)
Risks mitigated: Stateless misconception (assuming LLMs remember prior calls)Hallucinated actions from missing contextAssembly exhausting capacity needed for reasoning and output

Sources

Ch1.6: T. Nguyen, "Stateful Orchestration - Pitfalls, Integration, and Synthesis," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.6. ISBN: 9798244538229.
Ch3.5: T. Nguyen, "Prompt Optimization, Few-Shot Learning, Fine-Tuning," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 3.5. ISBN: 9798244538229.
Ch3.10: T. Nguyen, "Efficiency Metrics," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 3.10. ISBN: 9798244538229.
Ch5.7: T. Nguyen, "Episodic Memory," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.7. ISBN: 9798244538229.
Ch5.9: T. Nguyen, "Working Memory," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.9. ISBN: 9798244538229.
Ch8.3: T. Nguyen, "Token Economics and Architecture," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 8.3. ISBN: 9798244538229.
Ch10.1: T. Nguyen, "Conversational UI," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.1. ISBN: 9798244538229.