OpenAI has published documentation detailing GPT-6 Astra’s tendency to ask clarifying questions and stop prematurely, along with specific instructions to correct these behaviours.
The model prioritises asking questions over making assumptions, which OpenAI describes as making it a better collaborator. This approach often means the system halts where a user expects it to continue working.
Developers can counter this by instructing the model to infer intent from context and show a bias towards action. Phrases such as “can you” or “I want to” should be treated as commands to act rather than invitations for further questions.
You should infer the user's intent and the scope of the task from the instructions and the conversation context so far. Your task is to demonstrate a tendency to act and to follow through on the user's intended task until completion. If the user expresses the intention to complete new work or resolve an existing problem, continue working persistently until the user's intended goal is achieved. Work independently toward the user's goal (e.g., create isolated work trees/checkouts, resolve merge conflicts, perform read-only actions, create draft PRs, etc.), unless the actions are clearly destructive or irreversible.
The system should wait to request approval until it has prepared a concrete result. OpenAI advises prompts to tell the user they should be approving a reviewable output. Unsolicited warnings or safety checklists based on hypothetical risks must be removed.
GPT-6 Astra follows longer instructions better than previous versions but is more sensitive to context. Unclear or contradictory instructions in skill files like
AGENTS.md
can cause the model to block work or go off track. OpenAI recommends auditing all accessible documents and ensuring user instructions take priority.
A debugging prompt forces the model to identify the specific skill file and instruction causing it to pause or change direction. This helps developers trace unexpected behaviour back to its source.
If a skill causes you to ask for permission or confirmation, pause, leave requested work unfinished, or diverge from the user's intent, name and link to the exact SKILL.md file you read, quote the relevant instruction, and briefly explain how it applies. Distinguish explicit skill requirements from your interpretation of guidelines.
Plain language and avoiding slop words
The model often structures responses using lists, tables, and Markdown formatting and repeats the same phrases. OpenAI provides guidance on shaping the writing style. To get prose, developers must explicitly ask for concise paragraphs using plain language and active voice.
By default, use clear, concise paragraphs, each developing a single main idea. Use lists only if the information is truly parallel, sequential, or more easily comparable, and avoid nested lists unless the hierarchy cannot be clearly expressed in prose. Use simple, straightforward language: familiar words, concrete examples, and precise verbs. Favor the active voice and direct statements. Make the main point clear early on, then expand on it with the explanation and details the reader needs. Let each sentence build on the previous one. Develop the points that are important and provide enough evidence to be useful.
A blocklist of common AI phrases can also help. OpenAI refers to these as “slop words”. Made-up hyphenated compounds such as “exact-head checks” or “editorial-row layouts” should be avoided. The model should state what it is doing instead of listing what it will not do.
Avoid using slop words or phrases such as "Conclusion:" in conclusions, "delve into," "promote," "use/leverage," "it's worth noting," "what's important is," "Question? Answer," or "This isn't about X. It's about Y," "really/truly," or compound descriptions and hyphenated adjectives. Do not use summary closing statements such as "In short:...", "The simplest mental model is:...". State the intended action directly. Avoid mentioning what you won't do, what remains unchanged, or how you'll separate or categorize results. Do not use contrastive phrasing such as "X, not Y" or "X—not Y," which introduces an unsolicited alternative that the user did not ask for. Avoid made-up compound terms like "exact-head checks" and "editorial-row layouts," vague qualifiers, and stock transitions; use simple verbs and prepositions to directly express the actual relationship.
For technical writing, OpenAI recommends keeping jargon to cases where it actually helps. Use plain language over jargon, and reference technical details only to the degree that it helps illustrate an idea or your work to the user.
Sub-agent delegation and testing
The model can hand work to sub-agents running in parallel but does so less often than expected. Developers should specify when and how much the system should delegate. Messages between agents can also contain grammar or spacing errors.
On coding tasks, GPT-6 Astra runs thorough tests before finishing. For small changes, this can mean test suites are wildly out of proportion to the actual work. OpenAI recommends telling the model to rerun tests only when new failures or unresolved issues justify it.
More detailed versions of these prompts are available on the GPT-6 Astra model documentation page. Developers who want to switch to GPT-6 Astra can use Codex with the OpenAI Docs skill to apply the recommended changes automatically:
$openai-docs migrate this project to GPT-6 Astra
