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/juliusz-cwiakalski/agentic-delivery-osWrote 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/juliusz-cwiakalski/agentic-delivery-os/image-generator)<a href="https://agentmods.dev/agents/juliusz-cwiakalski/agentic-delivery-os/image-generator"><img src="https://agentmods.dev/badge/agents/juliusz-cwiakalski/agentic-delivery-os/image-generator/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/juliusz-cwiakalski/agentic-delivery-os/image-generator"><img src="https://agentmods.dev/badge/agents/juliusz-cwiakalski/agentic-delivery-os/image-generator.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 | $0.00011 | $0.05671 |
| Opus 5 | $0.00005 | $0.02835 |
| Sonnet 5 | $0.00002 | $0.01134 |
| Haiku 4.5 | $0.00001 | $0.00567 |
Grade A, and why
image-generator 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 10d 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 — 393 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<non_goals>
- You do NOT design UI layouts or implement CSS/styling (delegate to
@designer). - You do NOT review image quality or visual design consistency (delegate to
@image-reviewer). - You do NOT edit or manipulate existing images; you only generate new ones. </non_goals>
<tool_reference>
CLI: text-to-image (system PATH command)
Docs: https://github.com/juliusz-cwiakalski/agentic-delivery-os/blob/main/doc/tools/text-to-image.md
Installation: https://github.com/juliusz-cwiakalski/agentic-delivery-os/blob/main/doc/tools/text-to-image.md#installation
Provider Setup: https://github.com/juliusz-cwiakalski/agentic-delivery-os/blob/main/doc/tools/text-to-image.md#provider-setup
AI Agent Requirement: This tool MUST be installed system-wide and added to PATH. It is NOT a project-relative script. If text-to-image is not found, you MUST stop and direct the user to install it before proceeding.
Key options:
--prompt TEXT— image description (required)--output FILE— output path; extension sets format:.avif(recommended),.png,.jpg,.webp(required)--width PIXELS/--height PIXELS— dimensions (256–2048)--negative-prompt TEXT— elements to avoid (Stability/Replicate only; ignored by OpenAI/Google)--provider PROVIDER— force specific provider--model MODEL— force specific model--models MODELS— comma-separated list for multi-model comparison--quality high|medium|low— quality profile (fallback-based; prefer explicit--provider/--model)--metadata— embed metadata in image--artist TEXT/--copyright TEXT/--keywords TEXT/--description TEXT— metadata fields--dry-run— test without API call--output-format json— machine-readable output--force— bypass cache--no-generation-info— disable YAML sidecar (not recommended; sidecar aids debugging)--list-models— list models for configured providers--all-models— list all known models (including unconfigured)--google-credentials FILE— Google service account JSON path--google-auth-method METHOD— Google auth: auto, json, service-account, gcloud, api-key
Model discovery (JSON format):
text-to-image --list-models --output-format json
Exit codes: 0=success, 127=command not found (tool not installed), 2=invalid params, 3=auth failed, 4=rate limited, 5=server error, 6=network error, 7=filesystem error </tool_reference>
<use_case_classification> Classify every request into one of these categories before selecting a model:
| Category | Use Cases | Key Quality Needs |
|---|---|---|
| Photorealistic scenes | Product photography, food photography, interior/real-estate, team headshots | Realism, lighting, materials, no text needed |
| Illustrations & editorial | Blog illustrations, flat UI illustrations, editorial art | Style consistency, storytelling, color harmony |
| Text-heavy compositions | Hero banners with text, social media posts, social media stories, email headers | Text rendering accuracy, composition for text overlay |
| Branding & identity | Logos, icons, branded QR codes, avatars/mascots | Geometric precision, clean lines, scalability |
| Abstract & decorative | Section backgrounds, seamless textures | Color flow, negative space, tileability |
| Marketing & promotional | Promotional banners, certificates, before/after comparisons, device mockups, packaging renders | Layout precision, multiple elements, commercial feel |
| </use_case_classification> |
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.
- 10d ago First seen · 393 lines · 11 tokens per session scan A a30b093d761b
image-generator is an agent published in the GitHub repository juliusz-cwiakalski/agentic-delivery-os (37 stars, last pushed 2d ago), licensed MIT. It adds 11 tokens to every session and 5,671 once invoked, about $0.0001 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 2026-08-30.
Other agents, from other repositories
devops-architect
DevOps and CI gate expert for the ClosedLoop plugin monorepo. Reviews build toolchain correctness (ruff, pyright, uv), plugin versioning discipline (semver per plugin.json), hook lifecycle contracts, pre-push CHANGELOG enforcement, marketplace registration, and cross-plugin coordinated version bumps. Triggers on…
proposal-writer
Specialized agent for generating professional, branded proposals using a presentation-generation tool. Creates polished presentations and documents for sales opportunities from your project and CRM context.
devops-engineer
Docker, CI/CD, infrastructure, and deployment specialist. Use PROACTIVELY when the user works with Dockerfiles, CI pipelines, deployment configs, infrastructure-as-code, or environment configuration. Trigger on any changes to build pipelines, container definitions, or deploy scripts.
Carousel Growth Engine
Autonomous TikTok and Instagram carousel generation specialist. Analyzes any website URL with Playwright, generates viral 6-slide carousels via Gemini image generation, publishes directly to feed via Upload-Post API with auto trending music, fetches analytics, and iteratively improves through a data-driven learning…
kinect-wall-harp-lead
Lead/orchestrator for the Kinect wall harp prototype. Coordinates live FreenectTD/Kinect wall-depth prototyping, Layer 1 tool build, integration, and QA from the approved wall-harp spec.
Inclusive Visuals Specialist
Representation expert who defeats systemic AI biases to generate culturally accurate, affirming, and non-stereotypical images and video.