Conclusion
We evaluate the impact of context files on coding agent performance for four common coding agents
on SWE-BENCH and the novel CTXBENCH, built from recent GitHub issues and less popular
repositories containing developer-written context files. We find that all context files consistently
increase the cost and number of steps required to complete tasks. LLM-generated context files have
a marginal negative effect on task success rates, while developer-written ones provide a marginal
performance gain, neither statistically significant.
Our trace analyses show that instructions in context files are generally followed and lead to more test-
ing and broader exploration; however, they do not function as effective repository overviews. Over-
all, our results suggest that context files don’t improve coding agent performance, and should only
contain specific additional instructions beyond what is already available in the codebase. This high-
lights a concrete gap between current agent-developer recommendations and observed outcomes,
and motivates future work on principled ways to automatically generate concise, task-relevant guid-
ance for coding agents.
Agents md file doesn’t actually do anything except using more tokens https://arxiv.org/pdf/2602.11988
I feel like you either didn’t read your article or didn’t read the post article
To quote the research paper conclusion…
Conclusion We evaluate the impact of context files on coding agent performance for four common coding agents on SWE-BENCH and the novel CTXBENCH, built from recent GitHub issues and less popular repositories containing developer-written context files. We find that all context files consistently increase the cost and number of steps required to complete tasks. LLM-generated context files have a marginal negative effect on task success rates, while developer-written ones provide a marginal performance gain, neither statistically significant. Our trace analyses show that instructions in context files are generally followed and lead to more test- ing and broader exploration; however, they do not function as effective repository overviews. Over- all, our results suggest that context files don’t improve coding agent performance, and should only contain specific additional instructions beyond what is already available in the codebase. This high- lights a concrete gap between current agent-developer recommendations and observed outcomes, and motivates future work on principled ways to automatically generate concise, task-relevant guid- ance for coding agents.
That sounds like “Agents.MD doesn’t work” to me.
Am I missing something?