Paper: Dev-Primitives turn repo artifacts into active agents with resident LLMs
alex_verem · x · 2026-10-12
Source link for the HERMES thread (arXiv:2610.07832, by Haibo Jin et al.). The paper proposes Dev-Primitives, a modular executable abstraction that turns repository components from passive artifacts into active participants: each primitive pairs an artifact with a resident LLM providing natural-language reasoning, inter-component communication, and localized self-modification grounded in its own implementation. On top of this, HERMES instantiates the primitives at repository scale via dependency-aware dynamic activation and a bug-diagnosis mechanism mapping execution evidence back to the components that need revision, addressing context explosion and semantic drift in long-horizon agentic workflows.
Related event: HERMES assigns a resident AI model to every code file(2 posts)→
More from coding & agent
- Claude-driven Blender 3D modeling from images makes this dev bullish on inspection gear — jwt0625 · 2026-10-12
- Work2Skill Turns Human GUI Demos into Reusable Agent Skills, Beating No-Skill Baselines — nex-agi · 2026-10-12
- UCSD paper bakes theory of mind into RL reward, letting a 7B agent model partners solo — rohanpaul_ai · 2026-10-12
- Indie maker shows Claude Code prompt that builds SEO free tools, driving 4M impressions — tibo_maker · 2026-10-12
- Codex and ChatGPT Work Hit 40M Weekly Active Users, On Track for 100M by March 2027 — FinanceYF5 · 2026-10-12
- Two misconceptions in the REA era: software isn't dead, coding is — AlchainHust · 2026-10-12