Dev: 'Showing your prompt' is now meaningless — context is references, skills and examples
trq212 · x · 2026-09-29
Developer trq212 observes that the key asset in AI coding workflows is no longer a single prompt but the surrounding context engineering: reference repos, skills, examples, web searches, even other AI APIs. He routinely has his agent read three of his own repos, search the web for references, and call external APIs before producing results — making 'sharing your prompt' essentially meaningless.
More from coding & agent
- Claude Code skips RAG and just uses grep — simplicity wins again — srchvrs · 2026-09-29
- NVIDIA moves agent safety into the infrastructure, argues Ben Bajarin's new research note — BenBajarin · 2026-09-29
- Mo debuts AI QA engineer running 100+ parallel agents, claims 3.8x more bugs than Codex — ycombinator · 2026-09-29
- MotherDuck's prompt_jev() sees surprise production adoption one week after launch — hackgoofer · 2026-09-29
- Agent-written code looks worse but ships with fewer bugs than human builds, devs say — intellectronica · 2026-09-29
- Building evidence-backed sales deal recommendations on top of Hindsight agent memory — Best_Fox_3488 · 2026-09-29