# Project Prometheus: Bridging the Intent Gap in Agentic Program Repair via Reverse-Engineered Executable Specifications

arXiv: https://arxiv.org/abs/2604.17464 · HTML: https://mlsanity.com/paper/2604.17464/

## 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 7 of its 18 references that are in the map.
- First posted on arXiv 19 Apr 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, 7 of 7)

- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [RepairAgent: An Autonomous, LLM-Based Agent for Program Repair](https://mlsanity.com/paper/2403.17134/) (2024)
- [Less Training, More Repairing Please: Revisiting Automated Program Repair via Zero-shot Learning](https://mlsanity.com/paper/2207.08281/) (2022)
- [SpecRover: Code Intent Extraction via LLMs](https://mlsanity.com/paper/2408.02232/) (2024)
- [Hybrid Automated Program Repair by Combining Large Language Models and Program Analysis](https://mlsanity.com/paper/2406.00992/) (2024)
- [Large Language Models are Qualified Benchmark Builders: Rebuilding Pre-Training Datasets for Advancing Code Intelligence Tasks](https://mlsanity.com/paper/2504.19444/) (2025)
- [Abstain and Validate: A Dual-LLM Policy for Reducing Noise in Agentic Program Repair](https://mlsanity.com/paper/2510.03217/) (2025)

## Abstract (arXiv)

The transition from neural machine translation to agentic workflows has revolutionized Automated Program Repair (APR). However, existing agents, despite their advanced reasoning capabilities, frequently suffer from the ``Intent Gap'' -- the misalignment between the generated patch and the developer's original intent. Current solutions relying on natural language summaries or adversarial sampling often fail to provide the deterministic constraints required for surgical repairs. In this paper, we introduce textscPrometheus, a novel framework that bridges this gap by prioritizing textitSpecification Inference over code generation. We employ Behavior-Driven Development (BDD) as an executable contract, utilizing a multi-agent architecture to reverse-engineer Gherkin specifications from runtime failure reports. To resolve the ``Hallucination of Intent,'' we propose a Requirement Quality Assurance (RQA) Loop, a mechanism that leverages ground-truth code as a proxy oracle to validate inferred specifications. We evaluated textscPrometheus on 680 defects from the Defects4J benchmark. The results are transformative: our framework achieved a total correct patch rate of 93.97% (639/680). More significantly, it demonstrated a Rescue Rate of 74.4%, successfully repairing 119 complex bugs that a strong blind agent failed to resolve. Qualitative analysis reveals that explicit intent guides agents away from structurally invasive over-engineering toward precise, minimal corrections. Our findings suggest that the future of APR lies not in larger models, but in the capability to align code with verified, Executable Specifications -- whether pre-existing or reverse-engineered.