Borrowing it
Nothing to install: this file belongs to Dynokostya/just-works. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Dynokostya/just-works/main/.claude/skills/gpt-5-6-prompting/SKILL.mdgit clone --depth 1 https://github.com/Dynokostya/just-worksWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/dynokostya/just-works/gpt-5-6-prompting)<a href="https://agentmods.dev/skills/dynokostya/just-works/gpt-5-6-prompting"><img src="https://agentmods.dev/badge/skills/dynokostya/just-works/gpt-5-6-prompting/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/dynokostya/just-works/gpt-5-6-prompting"><img src="https://agentmods.dev/badge/skills/dynokostya/just-works/gpt-5-6-prompting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 148 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00135 | $0.04800 |
| Opus 5 | $0.00068 | $0.02400 |
| Sonnet 5 | $0.00027 | $0.00960 |
| Haiku 4.5 | $0.00014 | $0.00480 |
Grade A, and why
gpt-5-6-prompting scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 309 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GPT-5.6 Prompt Writing Guidelines
When to Use
- Creating or editing prompts targeting GPT-5.6 (any variant:
gpt-5.6-sol,gpt-5.6-terra,gpt-5.6-luna) - Calibrating reasoning effort, verbosity, autonomy boundaries, and tool routing for GPT-5.6 workloads
- Migrating prompt text from GPT-5.5, GPT-5.4, GPT-5.3-Codex, or older GPT models
- Diagnosing 5.6-specific behaviors (concise-by-default output, instruction-conflict instability,
mediumdefault reasoning, proactive multi-step execution)
Overview
GPT-5.6 is OpenAI's frontier family. gpt-5.6-sol is the flagship (the bare gpt-5.6 alias routes to it); gpt-5.6-terra balances cost; gpt-5.6-luna targets high-volume efficiency. Sol and Terra run ~1.05M-token context with 128K max output; Luna runs 400K context, 128K max output.
Compared with GPT-5.5, it reaches frontier performance with fewer output tokens, is more concise by default, follows prompt contracts more tightly (so conflicting instructions create instability), executes multi-step work more proactively, and has stronger layout and design judgment. New capabilities relevant to prompt design: programmatic tool calling, persisted reasoning across turns, pro mode for quality-first work, and multi-agent coordination (beta).
The core discipline is lean prompting: OpenAI measured 10-15% eval-score improvement with 41-66% token reduction from pruning prompts — GPT-5.6 rewards removing scaffolding more than adding it.
- Outcome-first: Strongest when the prompt defines destination, constraints, evidence, and completion bar, then leaves the path to the model.
- Tight contract-following: Follows prompt contracts closely; duplicated or conflicting instructions destabilize behavior. State each instruction once.
- Concise by default: More concise than GPT-5.5 — carried-over brevity blocks can now cut content you need. Define what brief answers must include.
- Proactive and persistent: Carries multi-step tasks forward on its own; needs approval boundaries, not step-by-step supervision.
- Strong planning over tools: Needs less fallback and invocation scaffolding than 5.5; still benefits from explicit prerequisite-retrieval and routing rules.
- Stronger design judgment: Better layout, hierarchy, and visual taste — constrain it to the existing design system rather than prescribing layout steps.
- Legacy-prompt penalty: Process-heavy stacks carried from older models narrow the search space and waste tokens.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 309 lines · 135 tokens per session scan A 4bdeaac6cc9a
gpt-5-6-prompting is a skill published in the GitHub repository Dynokostya/just-works (14 stars, last pushed 3d ago), licensed Apache-2.0. It adds 135 tokens to every session and 4,800 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
llm-app-patterns
Production-ready patterns for building LLM applications. Covers RAG pipelines, agent architectures, prompt IDEs, and LLMOps monitoring. Use when designing AI applications, implementing RAG, building agents, or setting up LLM observability.
prompt-optimization
Improve a prompt on the evaluations workbench through a measured loop. Score the baseline first, then duplicate the target column, form a hypothesis from failing rows, edit the copy's prompt draft, run, compare pass rate and cost, and repeat until the numbers hold. Use when the user asks to optimize or improve a…
enhance-prompt
Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.
prompt-engineer
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot…
seedance-vocab-en
This skill should be used when an English Seedance 2.0 prompt needs clearer production wording, less generic prose, or precise vocabulary for camera, lighting, motion, VFX, audio, and constraints. Route blocked prompts through seedance-filter for context and boundary review.
ideogram4
Prompting patterns for Ideogram 4 text-to-image — best-in-class in-image text rendering and exact color/layout control via structured JSON captions. Use when generating images that need legible on-image text (title cards, thumbnails, logos, signage, CTAs), precise brand colors, or controlled spatial layout. Triggers…