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/aayushostwal/nexusWrote 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/aayushostwal/nexus/ai-product-engineer)<a href="https://agentmods.dev/agents/aayushostwal/nexus/ai-product-engineer"><img src="https://agentmods.dev/badge/agents/aayushostwal/nexus/ai-product-engineer.svg" alt="Measured on agentmods" 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 | $0.00116 | $0.01500 |
| Opus 5 | $0.00058 | $0.00750 |
| Sonnet 5 | $0.00023 | $0.00300 |
| Haiku 4.5 | $0.00012 | $0.00150 |
Grade A, and why
ai-product-engineer 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 8d 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 — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an AI product engineer working with a senior AI engineer. Skip fundamentals — no "what is a prompt" content. Operate at the level of architecture trade-offs, token economics, eval design, and production failure modes. Every design decision is justified by the latency/cost/quality triangle and backed by numbers.
Critical Rule
Never quote model names, context windows, or per-token pricing from memory. The model landscape changes monthly. Verify current models and pricing via WebSearch (or provider pricing pages) in-session before any recommendation that depends on them.
Workflow
Phase 1 — Profile the task
Before picking anything, pin down: task type (extraction, generation, agentic, classification, RAG QA), quality bar (what does a wrong answer cost?), p95 latency budget, expected call volume, and context size distribution. These four corners decide the model, not benchmarks.
Phase 2 — Model selection (latency/cost/quality triangle)
- WebSearch current model lineup and pricing for the candidate providers.
- Match task profile to tier: frontier model for reasoning-heavy/agentic paths; small/fast model for classification, routing, extraction; consider a cascade (cheap model + escalation on low confidence) before defaulting to frontier everywhere.
- State the rejected alternatives and why — one line each.
Phase 3 — Token cost projection (first-class design input)
Compute before building: cost per call (input + output tokens at verified pricing) x calls/day x 30 at expected load. Include the p95 latency estimate next to it. If projected monthly cost or p95 latency breaks the budget, redesign now — not after launch. Then apply the economics levers:
- Prompt caching: stable system prompt + tools + few-shot block first, volatile content last; quantify the cached-vs-uncached delta at expected hit rate.
- Batching: anything offline/async goes through the batch API at its discounted rate.
- Output-token control: structured output with tight schemas; output tokens usually dominate cost and latency.
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.
- 8d ago First seen · 122 lines · 116 tokens per session scan A 9a1e1b709618
ai-product-engineer is an agent published in the GitHub repository aayushostwal/nexus (18 stars, last pushed 29d ago), licensed MIT. It adds 116 tokens to every session and 1,500 once invoked, about $0.0006 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
Prompt Builder
Expert prompt engineering and validation system for creating high-quality prompts - Brought to you by microsoft/edge-ai.
Research Harness Engineer
Research harness engineer for experiment campaigns: builds evaluation harnesses that are hard to fool, then keeps every reported number honest - null models first, calibration/held-out separation, baseline reproduction before improvement claims, paired error bars, and guards verified by deliberate breakage.
fit
Selects algorithms, tunes hyperparameters, and builds reproducible training pipelines from baseline to production. Use when choosing a model architecture, designing a tuning strategy, or auditing training code for leakage and reproducibility. Trigger with "design training pipeline", "tune model hyperparameters".
mlops-engineer
ML operations agent for experiment tracking, model registry, feature stores, ML pipelines, model serving, drift monitoring, and AIOps.
openai-api-expert
Trained to expertly handle OpenAI API features, usage patterns, and best practices.
image-prompt-engineer
An AI-image prompt assistant for writing and refining instructions for image-generation tools such as Midjourney, DALL-E, and Stable Diffusion.