codeprakhar25

Context Files Coding Agents — Development skill for Claude Code

Development community

Do Context Files Help Coding Agents.

How to install Context Files Coding Agents

This entry records only its repository, not the path inside it, so there is no exact command to give. Open codeprakhar25/context-files-coding-agents and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Context Files Coding Agents does

Do Context Files Help Coding Agents? Two-agent (Claude Code + Codex) ablation on whether AGENTS.md/CLAUDE.md change coding-agent correctness & efficiency.

Alternatives in Development

Benjamin Plus Skill — Benjamin-Plus: a measured token-efficiency skill for coding agents (−17.9% cost median, quality unchanged) 272 ★
FlowBoard — Local-first project management and persistent context for OpenClaw, Claude Code, Codex, Cursor, and multi-agen 94 ★
Nia — Nia is a context-augmentation layer for agents, primarily designed for coding agents 74 ★

README

Do Context Files Help Coding Agents?

**A two-agent ablation study on whether repository context files (`AGENTS.md` / `CLAUDE.md`) change how well LLM coding agents solve real GitHub issues.**

This is the code-and-data release for the paper *"Do Context Files Help Coding Agents? A Two-Agent Ablation Study on Real Repositories"* (REALM @ EMNLP 2026, under review). It contains the full experimental harness, the safety infrastructure, the benchmark task specifications, the statistical analysis scripts, and an aggregated per-cell results table.

TL;DR

We run two production coding agents (**Claude Code** and **OpenAI Codex CLI**) on real, merged pull-request tasks from three Python repositories, under three context-injection strategies, and measure whether the presence of a repository context file changes task **correctness** and **efficiency**.

**Headline finding — a correctness null, replicated across both agents.** Adding the repository's `AGENTS.md` (in full, or via a retrieval-style split) does **not** measurably change the probability that the agent's patch passes the gold PR tests.

Agent	`none`	`always_on`	`selective`
Claude Code (15 tasks)	53.3%	55.6%	55.6%
Codex (17 tasks)	58.8%	56.9%	52.9%

Within-task omnibus permutation test: p = 1.000 (Claude); no detectable effect on either agent.
TOST equivalence: effect bounded to ≤10pp (Claude) / ≤15pp (Codex).
The result is power-limited, and we say so: at n≈15–17 tasks × 3 repeats the minimum detectable effect is >30pp; detecting a 10pp effect at 80% power would need ~120 tasks. We report this as the binding constraint, not a proof of zero.

**Where context *did* move the needle — process efficiency, not outcome.** On the OpShin/opshin tasks, Claude Code with context ran the slow full test suite far less often (blind full-suite runs **3.67 → 2.44 → 1.67** across `none`→`always_on`→`selective`), cutting wall-clock time **