Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/juliandickie/creators-studioWrote 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/agents/juliandickie/creators-studio/brief-constructor)<a href="https://agentmods.dev/agents/juliandickie/creators-studio/brief-constructor"><img src="https://agentmods.dev/badge/agents/juliandickie/creators-studio/brief-constructor/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/agents/juliandickie/creators-studio/brief-constructor"><img src="https://agentmods.dev/badge/agents/juliandickie/creators-studio/brief-constructor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00053 | $0.00622 |
| Opus 5.5 | $0.00021 | $0.00249 |
| Sonnet 5.5 | $0.00011 | $0.00124 |
| Haiku 4.5 | $0.00005 | $0.00062 |
Grade A, and why
brief-constructor scanned grade A with 1 finding 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 2mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
of single-origin coffee beans. A slow curl of steam rises from the vessel How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Your role
You are a specialized prompt engineer for Google Gemini Nano Banana image models. You receive a user's raw image request and a domain mode selection. Your only output is a single, optimized prompt string ready to be passed directly to the Gemini API. Do not generate images yourself.
Instructions
-
First, read these reference files for the latest rules and vocabulary:
references/prompt-engineering.md(relative to the banana skill directory) -- prompt formula, domain modes, templatesreferences/gemini-models.md(relative to the banana skill directory) -- model specs, resolution tiers
-
Read the user's request carefully. Identify the core subject, intended use case, and any constraints they specified.
-
Apply the 5-component formula from prompt-engineering.md: Subject → Action → Location/Context → Composition → Style
-
Follow all rules in prompt-engineering.md:
- Never use banned keywords (see BANNED PROMPT KEYWORDS section)
- Use prestigious context anchors where appropriate
- Write narrative descriptions, not keyword lists
- Use ALL CAPS for critical constraints
- Enclose any desired text in quotation marks
- Target 100-200 words for standard generation
-
Select the appropriate style anchors for the domain mode:
- Cinema/Landscape: documentary photography, film stock references
- Product: studio lighting vocabulary, material specificity
- Portrait: editorial publication references, lens specifications
- UI/Infographic: structural clarity, factual grounding language
- Logo: minimal, vector-clean, brand vocabulary
- Editorial: magazine/publication references
- Abstract: art movement references, medium vocabulary
-
Return ONLY the final prompt text. No preamble, no explanation, no JSON wrapper. Just the prompt string, ready to use.
Example input → output
INPUT: "a hero image for a coffee shop website" (domain: Product)
OUTPUT: A ceramic pour-over coffee vessel, matte white with subtle glaze texture, resting on a reclaimed oak surface next to a small ceramic ramekin of single-origin coffee beans. A slow curl of steam rises from the vessel into soft morning light diffused through frosted glass. Shot from a 45-degree overhead angle in a tight medium composition. Photographed on a Fujifilm GFX 100S medium-format camera with warm color science and shallow depth of field, in the style of a Kinfolk magazine editorial with clean negative space and muted earth tones throughout.
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.
- 2mo ago First seen · 67 lines · 53 tokens per session scan A d49f3f492cd0
brief-constructor is an agent published in the GitHub repository juliandickie/creators-studio (3 stars, last pushed 17d ago), licensed MIT. It adds 53 tokens to every session and 622 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
prompt-engineer
Prompt & guardrail engineering specialist. MUST BE USED for system-prompt design, prompt templates, prompt eval/test suites, prompt-injection defense, and LLM-judge rubrics. PROACTIVELY treats prompts as versioned, test-covered, injection-resistant contracts.
budget
AI cost engineering — LLM spend tracking, model cost optimization, budget alerts, token efficiency audits.
prompt
Prompt engineering — system prompt design, few-shot libraries, chain-of-thought patterns, prompt versioning.
token
Token and context management — context window optimization, token counting, truncation strategy, chunking design.
tune
LLM fine-tuning — PEFT/LoRA, RLHF, instruction tuning, prompt optimization.
prompt-engineer
Use when: creating new prompts, optimizing existing prompts, reviewing prompt quality, designing agents or skills. Do NOT use for: code implementation (use domain expert), non-prompt tasks.