AILANTA
← All research
GitHub · Agent infrastructure

AI agent infrastructure moves from orchestration to control

An expanded first-half sample shows that infrastructure already dominates the agent ecosystem. The next shift is happening inside it: from general orchestration toward memory, observability, and reliability.

15.07.20265,964 GitHub repositories3,467 agent projectsfrozen snapshot
Main finding

Not a new infrastructure wave, but specialization inside an established layer

The infrastructure share barely moved: 84.4% in Q1 and 85.5% in Q2. Yet memory and context gained 1.7 pp, observability and evaluation gained 1.0 pp, and recovery and reliability gained 0.4 pp, while general orchestration declined by 2.2 pp.

01

Sample scale and maturity

Quarterly cohorts after repository deduplication.

Agent projects1,7961,671

Q1 → Q2

Infrastructure share84.4% → 85.5%

95% CI: 83.887.1%

GitHub Releases present79% → 71%

top 100 per quarter

Verified on PyPI9% → 11%

exact project URL match

02

What is changing inside the stack

Each project belongs to one primary layer to avoid double counting.

Q1 2026Q2 2026
Orchestration-2.3 pp
58.6%56.3%
Agent applications-1.1 pp
15.6%14.5%
Memory and context+1.8 pp
10%11.8%
Sandboxing and execution+0.1 pp
9.1%9.2%
Observability and evaluation+1 pp
4.5%5.5%
Identity and permissions+0.1 pp
1.5%1.6%
Recovery and reliability+0.5 pp
0.7%1.2%

Orchestration remains the largest segment, but its relative share is shrinking. Growth in narrower control functions points from assembling agent workflows toward operating them in long-running and critical processes.

03

Three market implications

01

Memory becomes infrastructure

Context, retrieval, and long-term memory are moving from application features into a reusable layer of agent systems.

02

Quality requires observability

Growth in evals, tracing, and monitoring reflects a new bottleneck: agent value depends on reproducible production behavior, not demos.

03

Reliability is small but accelerating

Recovery represents only 1.2%, but it is growing from a small base. This is an early market, not an established one.

04

Supporting evidence

These sources support interpretation but do not replace the primary GitHub statistics.

41current Show HN launches matched the agent taxonomy
OpenAlexshows sustained academic activity across the principal control layers
Methodology and limitations

GitHub: new public repositories from January–June 2026, combining stored broad cohorts with targeted searches for agent, agentic, MCP, multi-agent, computer-use, and tool-calling. Repositories are deduplicated by ID. Classification uses names, descriptions, and topics. Maturity is checked only for the top 100 agent projects in each quarter. Product Hunt is excluded.

Limitations: stars introduce survivorship bias; OpenAlex searches overlap and may lag; PyPI verification requires an exact GitHub project link; the official Show HN API only provides a current snapshot. The change in overall infrastructure share sits within overlapping confidence intervals and does not by itself establish growth.