# SpecFirst: Behavioral Specification Elicitation as a First-Class Step in Agent-Based Program Synthesis from Scratch

arXiv: https://arxiv.org/abs/2607.27167 · HTML: https://mlsanity.com/paper/2607.27167/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 21 of its 39 references that are in the map.
- First posted on arXiv 29 Jul 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 21 of 21)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis](https://mlsanity.com/paper/2203.13474/) (2022)
- [StarCoder: may the source be with you!](https://mlsanity.com/paper/2305.06161/) (2023)
- [LLMs Get Lost In Multi-Turn Conversation](https://mlsanity.com/paper/2505.06120/) (2025)
- [Unsupervised Translation of Programming Languages](https://mlsanity.com/paper/2006.03511/) (2020)
- [RobustFill: Neural Program Learning under Noisy I/O](https://mlsanity.com/paper/1703.07469/) (2017)
- [Trae Agent: An LLM-based Agent for Software Engineering with Test-time Scaling](https://mlsanity.com/paper/2507.23370/) (2025)
- [ProgramBench: Can Language Models Rebuild Programs From Scratch?](https://mlsanity.com/paper/2605.03546/) (2026)
- [LLM4Decompile: Decompiling Binary Code with Large Language Models](https://mlsanity.com/paper/2403.05286/) (2024)
- [Drift No More? Context Equilibria in Multi-Turn LLM Interactions](https://mlsanity.com/paper/2510.07777/) (2025)
- [Lingxi: Repository-Level Issue Resolution Framework Enhanced by Procedural Knowledge Guided Scaling](https://mlsanity.com/paper/2510.11838/) (2025)
- [Solving Context Window Overflow in AI Agents](https://mlsanity.com/paper/2511.22729/) (2025)
- [Quantifying Laziness, Decoding Suboptimality, and Context Degradation in Large Language Models](https://mlsanity.com/paper/2512.20662/) (2025)

## Abstract (arXiv)

LLM-based agents excel at software engineering tasks where an existing codebase provides context, but constructing a program from scratch remains fundamentally harder. Recent benchmarks such as ProgramBench quantify this gap: given only natural-language documentation and an execute-only binary as a behavioral oracle, even frontier models solve fewer than 1% of instances. Existing frameworks conflate documentation reading, behavioral exploration, and code synthesis into a single pass, causing agents to probe insufficiently, lose behavioral intent as context drifts, and propagate early misinterpretations into the final implementation. Inspired by classical requirements engineering, we argue that behavioral specification elicitation should be a first-class phase that precedes implementation. We present SpecFirst, a two-stage framework that forces the specification elicitation before code synthesis. A dedicated spec agent first probes the binary and combines observations with documentation into a structured specification. Next, a code synthesis agent then uses this specification to drive implementation. This decomposition resolves documentation ambiguities before coding begins and provides a stable behavioral reference throughout synthesis. We evaluate SpecFirst on all 200 ProgramBench instances across four models spanning two families and an order of magnitude of capability. SpecFirst consistently outperforms the single-loop baseline, improving test pass rates by 6.9%-21.3% and binary exploration coverage by 9.4%-18.5%, all statistically significant. Behavioral analysis on code synthesis further shows that a prior specification enables earlier and more sustained code construction. Our results demonstrate that an explicit requirements-engineering phase is an effective paradigm for from-scratch program construction.