Openai told developers on september 11 to strip their agents.md files for gpt-6 astra
openai told developers on september 11 to strip their agents.md files for gpt-6 astra: drop the pre-reads, the test nagging, the old prohibitions written to stop weaker models. guardrail prompts are now the slop.
the instructions that protect a model become the noise a better model has to ignore.
Context
OpenAI's developer blog post of 11 September 2026, Rethinking skills and prompts for GPT-6 Astra, by Eric Provencher of Codex, has the subtitle revisit skill descriptions, AGENTS.md and task prompts to avoid bloated context. It says telling the model to read several docs before every edit is a great way to burn context and recommends contextual pointers instead. It says previous models needed encouragement to run tests and GPT-6 Astra does that on its own, so the same instructions can lead to unnecessary testing, and that AGENTS.md can grant explicit permission to run safe local tests. For boundaries added to make older models ask first, it says to consider updating that language because Astra could take it too seriously and may stop work. It also says to define completion upfront.
The date and venue match, but nothing in the post says strip. It says revisit, shorten and make instructions contextual, and keep AGENTS.md current, not delete it. It does not say to remove safety boundaries, and the note's told developers to strip and drop old prohibitions is a stronger paraphrase. The advice is model-specific from one OpenAI author for one model, and does not show guardrail prompts are noise for other models. The Decoder's framing of fewer guardrails is the outlet's wording. Guardrail prompts are now the slop is the author's opinion, partly supported for pre-reads and test prompts, not for safety boundaries.
Related work
- GPT-6 Astra needs leaner prompts (The Decoder, 12 Sep 2026) ↗Secondary, snippet only.
Watch next
- OpenAI's own Astra prompting docs.
Sources
- Rethinking skills and prompts for GPT-6 Astra (OpenAI Developers, 11 Sep 2026)developers.openai.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 18 September 2026 at 19:16 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →