OpenAI rethinks skills and prompts for GPT-6 Astra; shared Codex audit prompt
daniel_mac8 · x · 2026-09-12
OpenAI's dev blog published "Rethinking skills and prompts for GPT-6 Astra", covering how to redesign skills, AGENTS.md, and decision boundaries for the new model. Daniel Mac8 shares a ready-to-use Codex prompt: have Codex read Eric Provencher's post, then audit your skills, AGENTS.md, and decision boundaries against its advice — "Astra-fying" your Codex setup.
Related event: OpenAI Publishes Guide on Rethinking Skills and Prompts for GPT-6 Astra(2 posts)→
More from coding & agent
- Kimi K2.7 RL recap: gains transfer to unseen benchmarks while steps drop ~35% — echen · 2026-09-12
- RL on just 1,700 tasks lifts Kimi K2.7 across five coding benchmarks — echen · 2026-09-12
- Expected Parrot's universal remote cache hits 30M entries to fix AI-agent reproducibility — soumitrashukla9 · 2026-09-12
- iLands Agents Obsess Over Fact-Checking, Burning Tokens on Mutual Suspicion — repligate · 2026-09-12
- mcp-search-proxy hides 532 MCP tools (~179k tokens/turn) behind 4 tools via BM25 — pisa_p · 2026-09-12
- DeepSeek v4.1-Flash multimodal model lands on Modal with 16B/8B asymmetric activation — charles_irl · 2026-09-12