# Context Engineering

Mastra released Knowledge, a shared store that organizes company documents, policies and decisions as named Markdown pages with scopes, giving agents company-wide context. Previously agents could get context from instructions, retrieval-augmented generation, skills, or tools that query APIs and databases, but those paths are usually read-only and need manual updates when information changes. Knowledge works with signals, Mastra's context engineering primitive, to bring relevant knowledge into an agent's context and keep it current while the agent runs, and agents can search, read and write pages within the scopes they can access.

Mastra released Knowledge, a shared store that organizes company documents, policies and decisions as named Markdown pages with scopes, giving agents company-wide context. Previously agents could get context from instructions, retrieval-augmented generation, skills, or tools that query APIs and databases, but those paths are usually read-only and need manual updates when information changes. Knowledge works with signals, Mastra's context engineering primitive, to bring relevant knowledge into an agent's context and keep it current while the agent runs, and agents can search, read and write pages within the scopes they can access.

LangChain's open source Deep Agents harness applies five context engineering techniques against context rot in long tasks: offloading oversized tool results and user messages to the filesystem, summarizing history near the limit, loading skills on demand, isolating noisy work in subagents, and keeping long-term memory as a plain file. Information is moved out of the active window rather than deleted and can be read back, so memory lives in storage the user controls and survives a model or provider switch.

LangChain's open source Deep Agents harness applies five context engineering techniques against context rot in long tasks: offloading oversized tool results and user messages to the filesystem, summarizing history near the limit, loading skills on demand, isolating noisy work in subagents, and keeping long-term memory as a plain file. Information is moved out of the active window rather than deleted and can be read back, so memory lives in storage the user controls and survives a model or provider switch.

The work introduces a unified formulation of context optimization that interprets an agent memory system update as an optimization update procedure over the model's context, and on that basis proposes GraphMemory, a lightweight graph-based memory that accumulates, refines, organizes, and connects reusable strategies; each query retrieves only the relevant subgraph, so under bounded retrieval the amount of retrieved memory stays constant as the number of processed examples grows, and experiments show competitive downstream performance while using approximately 81-85% fewer memory-construction tokens than the baselines.

The work introduces a unified formulation of context optimization that interprets an agent memory system update as an optimization update procedure over the model's context, and on that basis proposes GraphMemory, a lightweight graph-based memory that accumulates, refines, organizes, and connects reusable strategies; each query retrieves only the relevant subgraph, so under bounded retrieval the amount of retrieved memory stays constant as the number of processed examples grows, and experiments show competitive downstream performance while using approximately 81-85% fewer memory-construction tokens than the baselines.

This work builds and releases the GHAgentFiles dataset together with the open-source extraction tool cofee, identifying 27,717 repositories with coding agent files out of 165,281 candidates and extracting 142,294 context, skill and subagent file histories (401,873 file revisions across 185,795 commits, spanning 22 November 2022 to 1 July 2026), and it presents a preliminary observational analysis showing that coding agent files emerge around early 2025, peak in additions and modifications in early 2026, and that the generic AGENTS.md has become the most popular naming practice.

This work builds and releases the GHAgentFiles dataset together with the open-source extraction tool cofee, identifying 27,717 repositories with coding agent files out of 165,281 candidates and extracting 142,294 context, skill and subagent file histories (401,873 file revisions across 185,795 commits, spanning 22 November 2022 to 1 July 2026), and it presents a preliminary observational analysis showing that coding agent files emerge around early 2025, peak in additions and modifications in early 2026, and that the generic AGENTS.md has become the most popular naming practice.