Human-written AGENTS.md context files improve agent performance by approximately 4 per cent, while LLM-generated context files decrease performance by 3 per cent and increase token cost by over 20 per cent, as the article reports citing recent research¶
Claim · All claims
Evidence · 1 study · designr1·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Human-written context files improve agent performance by approximately 4 per cent, whereas LLM-generated context files decrease performance by 3 per cent and increase token cost by over 20 per cent. → Mikael Gorsky (Holon Institute of Technology) 2026
Evidence¶
Mikael Gorsky (Holon Institute of Technology) 2026¶
Mikael Gorsky (Holon Institute of Technology). (2026). ASE-26: a curriculum for agentic software engineering as a discipline. https://arxiv.org/abs/2606.01152 doi:10.48550/arxiv.2606.01152
q2 · i? · design · r1
The article reports this in Module 11's coverage of context engineering, citing two anonymous-authored arXiv preprints on the impact of AGENTS.md files; the article offers the percentages as reported support, not as a study conducted here.
"human-written context files improve agent performance by approximately 4 per cent, while LLM-generated context files decrease performance by 3 per cent and increase token cost by over 20 per cent"
Discussion¶
Related Claims¶
- May 2026 token telemetry was cache-dominant: 82.9% of 73,950,305 recorded tokens were cache reads, suggesting persistent workflows shift the economic unit from cost per token to cost per completed artifact — related
- Model selection experiments show newer models are not strict improvements, with component-specific effects on quality metrics — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
agents-md-human-written-context-files-improve-agent-performance |
title |
Human-written AGENTS.md context files improve agent performance by approximately 4 per cent, while LLM-generated context files decrease performance by 3 per cent and increase token cost by over 20 per cent, as the article reports citing recent research |
description |
Human-written AGENTS.md context files improve agent performance by approximately 4 per cent, while LLM-generated context files decrease performance by 3 per cent and increase token cost by over 20 per cent, as the art... |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — mikael-gorsky-holon-institute-of-technology-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.