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.
npx agentmods add skills/understudylabs/understudy-agent-tools/plan-hosted-runnpx skills add understudylabs/understudy-agent-tools --skill plan-hosted-rungit clone --depth 1 https://github.com/understudylabs/understudy-agent-toolsWhat 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 | $0.00089 | $0.03165 |
| Opus 5 | $0.00044 | $0.01582 |
| Sonnet 5 | $0.00018 | $0.00633 |
| Haiku 4.5 | $0.00009 | $0.00316 |
Grade A, and why
plan-hosted-run 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 3d 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 — 224 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plan Hosted Run
One skill for the single question "I want to run a hosted job — where, how long, and how much?" It has two halves that usually run together:
- Estimate — a defensible back-of-envelope for wall-clock and dollars,
local (Apple Silicon / MLX) versus cloud (rented GPU or serverless), with
every input labeled measured-or-assumed. Methodology below; constants,
benchmarks, and cited prices in
references/cost-estimation.md. - Route — match the job shape to the provider whose strengths actually
fit, with cited facts and honest caveats. Routing table below; per-provider
detail and citations in
references/providers.md.
This skill estimates and recommends. It never provisions, rents, or spends — those are explicit, separate user actions. Prices and features drift — verify on the provider's live pricing page before any spend.
Optimize the recommendation for reaching the workload objective, not for the
lowest sticker price. Present the recommended outcome-sized plan first, then a
cheaper diagnostic and a faster or higher-confidence option when useful. State
what each buys in capability, confidence, and time-to-answer; follow
../understudy/reference.md → Outcome-first spend
posture.
When to use
- "How long will fine-tuning / SFT / an RL run take on my Mac vs a rented GPU?"
- "What will it cost to generate N RL trajectories?" (policy rollouts at scale)
- "Where should I run this fine-tune / RL job / batch rollout?"
- "Who's cheapest/fastest for generating RL trajectories?"
- "I want managed GRPO vs I want to rent raw GPUs — who does which?"
- Sizing the spend before any hosted-RL handoff
(
../prepare-verifier-handoff/SKILL.md).
Not for: choosing a model
(../compare-model-sweep/SKILL.md), the
authenticated Understudy gateway
(../use-understudy-gateway/SKILL.md),
or running the job itself.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago First seen · 224 lines · 89 tokens per session scan A 5b391740b671
plan-hosted-run is a skill published in the GitHub repository understudylabs/understudy-agent-tools (16 stars, last pushed 4d ago), licensed MIT. It adds 89 tokens to every session and 3,165 once invoked, about $0.0004 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 skills, from other repositories
agent-platform-model-registry
Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.
bigquery-ai-ml
Leverages BigQuery's built-in machine learning and GenAI capabilities for advanced data analytics. Use when you need to write SQL queries that perform time-series forecasting, predict values, detect outliers or anomalies, find key drivers, perform semantic search or vector search, classify text, calculate similarity…
ondb
A logical analysis and reasoning tool for AI. Use when decomposing documents into structured knowledge, querying entities and relations, validating consistency, or indexing files. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", "analyze this document", entity CRUD, or cross-skill…
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
mixed-precision
Use FP16/BF16 mixed precision to accelerate training and reduce memory. Use when optimizing GPU performance.
sparse-autoencoder-training
Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.