Improving Claude Code Success Rates with RE-call Memory Layer

说明与用途

实验表明，与静态指令文件（CLAUDE.md）相比，在 Claude Code 中使用 RE-call 动态记忆层可将任务成功率提高 22.2%。静态文件将性能降低至 36.1%，而基础模型为 50.0%，这可能是由于噪声干扰所致。RE-call 提供了精确的上下文检索，但增加了令牌成本。

问题

Static instruction files (like CLAUDE.md) can introduce noise or conflicting constraints, leading to lower success rates in deterministic coding tasks compared to bare models. Agents need a way to retrieve relevant project history without polluting the context window.

结果

RE-call improved task success rates by 22.2 percentage points over static CLAUDE.md files and 8.3 points over bare Claude Code. The improvement was statistically significant (p=0.000145). Static files performed worse than bare models, likely due to noise. RE-call costs more per task but prevents expensive failure modes.

工作流

• Set up a temporary repository for each task.
• Define deterministic checkers to validate the final state of the repository.
• Run tasks using three configurations: Bare Claude Code, Claude Code with CLAUDE.md, and Claude Code with RE-call memory.
• Verify memory tool availability before counting sessions.
• Compare success rates and analyze paired comparisons to determine statistical significance.
• Review cost and token usage for each configuration.

证据

RE-call improved success over the CLAUDE.md baseline by 22.2 percentage points.

来源

The paired McNemar test gave a p-value of 0.000145.

来源

The static CLAUDE.md configuration performed worse than the bare configuration: Bare 50.0%, CLAUDE.md 36.1%.

来源