AILANTA
← All research
GitHub · Technical emergence

The agent wave is shifting from brands to practice

An expanded sample of new GitHub projects confirms the scale of agent development while revealing a more nuanced shift: the broad layer is stable, branded waves are fading, and portable skills and instruction quality are gaining share.

16.07.20266 months695 × 6 repositoriesfrozen v2
Main finding

Not continuous growth, but normalization after a March peak

ai-agent rose from 10.4% in January to 15.4% in March, then normalized to 11.2% in June—still above January. The ecosystem is not contracting; it is becoming less dependent on a single brand or protocol.

01

Six months of agent development

Author-provided topic share across all 695 repositories in each month.

0%4%8%12%16%JanFebMarAprMayJun
ai-agentclaude-codeagent-skillprompt-engineering
02

Topics changing share

GitHub topicJanuaryJuneChange
prompt-engineering0.7%2.4%+1.7 pp
local-first0.6%1.6%+1 pp
self-hosted1.4%2.4%+1 pp
image-generation0.4%1.4%+1 pp
ai-agent10.4%11.2%+0.8 pp
agent-skill4.2%4.9%+0.7 pp
claude-code12.1%8.5%-3.6 pp
openclaw4%0.7%-3.3 pp
mcp6.5%3.5%-3 pp
claude8.5%4.6%-3.9 pp
03

Three persistent shifts

01

Prompt engineering is rising again

Share rose from 0.7% to 2.4%. As agents become more complex, instructions and context management return as infrastructure problems.

02

Agent skills persist

agent-skill moved from 4.2% to 4.9% despite volatility, supporting a shift toward portable capabilities.

03

Branded themes lose concentration

Claude Code, Claude, MCP, and OpenClaw all lost share. Their use cases persist but diffuse across a broader tool stack.

04

Topic coverage

Cohorts are equal in size; metadata coverage is reported separately.

Jan
386 / 695
Feb
402 / 695
Mar
395 / 695
Apr
385 / 695
May
354 / 695
Jun
343 / 695