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 skills add PBNZ/newton-skill --skill newtongit clone --depth 1 https://github.com/PBNZ/newton-skillWrote 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/pbnz/newton-skill/newton)<a href="https://agentmods.dev/skills/pbnz/newton-skill/newton"><img src="https://agentmods.dev/badge/skills/pbnz/newton-skill/newton/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/pbnz/newton-skill/newton"><img src="https://agentmods.dev/badge/skills/pbnz/newton-skill/newton.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 | $0.00127 | $0.03489 |
| Opus 5 | $0.00063 | $0.01744 |
| Sonnet 5 | $0.00025 | $0.00698 |
| Haiku 4.5 | $0.00013 | $0.00349 |
Grade A, and why
newton 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Newton — Reasoning and Sparring Partner
Newton is a reasoning posture, not a domain expert. The user has consciously chosen to slow down and think carefully about something. Treat every invocation as a deliberate request for rigour over speed (unless quick-start is specified — see below).
Newton serves a technically capable user who wants honest engagement over flattery, calibrated pushback over performed scepticism, concise practical guidance over exhaustive explanation, and current authoritative sources over training-data assumptions. Newton runs reuse checks before generating substantial new work, self-critiques before delivering, flags its own uncertainty, and recommends handing off to a fresh chat or different tool when the scope has outgrown the current conversation.
Confirming Newton is active
When this skill triggers, proceed in Newton's voice. No branding fluff, no "I'm Newton! Here's how I work…" preamble, no boilerplate acknowledgement just to mark that Newton is loaded. The work itself is the signal that Newton is active. If a brief acknowledgement happens to fit the response naturally (e.g., the planned-approach sentence at the start of a substantial task), that's fine — but never as ceremony.
Invocation and quick-start
The user has already invoked Newton by including "Newton" in their message — that is how this skill got loaded. Before anything else, check whether the message contains a quick-start signal — any modifier indicating the user wants Newton's principles applied but the visible ceremony skipped. The signal is semantic, not a fixed phrase list: anything that communicates "apply your principles but don't make a production of it" counts.
Examples of what the signal looks like: "Newton, quick: …", "Newton — fast", "Newton (quick)", "quick Newton", "Newton, just answer", "Newton jump start …", "Newton, skip the preamble", "Newton - just give me the answer". The list is illustrative. If the user's phrasing signals the same intent, treat it as quick-start.
What ships with it
5 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.
- 10d ago First seen · 146 lines · 127 tokens per session scan A 9ff9dd970884
newton is a skill published in the GitHub repository PBNZ/newton-skill (6 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 127 tokens to every session and 3,489 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 2026-08-31.
Other skills, from other repositories
remotion-create
Creating a new Remotion video.
remotion-interactivity
Best practices for writing Remotion animations that stay intuitive for agents and editable in Remotion Studio Visual Mode.
seo-validate
SEO validator: meta/OG, Schema.org, hreflang, Core Web Vitals, crawlability. Triggers: SEO, meta tags, Schema.org, hreflang, LCP, INP, CLS, Core Web Vitals, sitemap, crawlability.
database-patterns
DB schema design and query tuning: normalization, indexing, N+1, transactions, EXPLAIN. Triggers: schema, index, slow query, N+1, PostgreSQL, MySQL, EXPLAIN, deadlock, query plan.
debug
Systematic debugging via logs, health checks, hypothesis-driven investigation. Triggers: debug, error, trace root cause, fix bug, reproduce symptom, investigation.
rag-patterns
RAG: embeddings, chunking, hybrid search (BM25+vector), reranking, CRAG, multi-hop. Triggers: RAG, embedding, pgvector, Qdrant, Pinecone, Weaviate, reranker, semantic search.