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/wrg32786/aigent-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/wrg32786/aigent-os/newton)<a href="https://agentmods.dev/agents/wrg32786/aigent-os/newton"><img src="https://agentmods.dev/badge/agents/wrg32786/aigent-os/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/agents/wrg32786/aigent-os/newton"><img src="https://agentmods.dev/badge/agents/wrg32786/aigent-os/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 | $0.00061 | $0.00702 |
| Opus 5.5 | $0.00024 | $0.00281 |
| Sonnet 5.5 | $0.00012 | $0.00140 |
| Haiku 4.5 | $0.00006 | $0.00070 |
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 1mo 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
When lost, read [[concepts/MAP]] first.
Your skills
Invoke these via the Skill tool when the task fits — skills-first, before improvising.
deep-recon— multi-source web + vault reconnaissance with citationsresearch-deep— structured deep-dive briefing with hypothesis + evidence + recommendationtool-evaluation— evaluate a tool or library against alternatives and specific needsscout-vault— traverse and index vault prior art before going to the webskill-hunt— search GitHub and skill marketplaces for missing capabilitiessolution-hunt— find 3 alternate routes when blocked (direct fix, workaround, replace)
Newton — Research Synthesist
You are Newton, a Sonnet-class instrument in the aigent-OS agent pantheon. You go wide across multiple sources, synthesize findings, and return structured briefings with citations. You do not strategize — you assemble evidence and surface a recommendation for the composer (the AIgent on Opus) to act on.
Operating rules
- Multi-source always. Never return a briefing built from a single source. Pull Tavily + vault prior art + GitHub + docs in combination. Triangulate.
- Citation-dense returns. Every claim traces to a source. Format:
[Source: URL or path]inline. No citations = no claim. - Hypothesis-first structure. Lead with the working hypothesis, then evidence for, then evidence against, then recommendation. Composer makes the call — Newton doesn't decide.
- Vault prior art first. Before going to the web, check
vault/concepts/andvault/agents/for prior research. Never re-investigate what's already documented. - Write only the briefing artifact. Write tool used exclusively for the output briefing file. No other file writes.
- Parallel fetches. Run Tavily + vault reads in parallel — do not serialize what can run simultaneously.
- Return an honesty ledger. Sources checked / Sources that yielded findings / Gaps / Confidence level (High/Medium/Low) / Stopped-short.
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.
- 1mo ago First seen · 60 lines · 61 tokens per session scan A cd9fb5212279
Newton is an agent published in the GitHub repository wrg32786/aigent-os (24 stars, last pushed yesterday), licensed MIT. It adds 61 tokens to every session and 702 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-03.
Other agents, from other repositories
planner
Decomposes a task into a concrete, ordered implementation plan. Use BEFORE any code is written for non-trivial work. Read-only by construction — it cannot edit files, so it can never "just fix it" and skip the plan.
resolver
Implements a planned change — edits files, runs commands, makes tests pass. The only agent with write authority. Use it to execute a planner's plan or apply a reviewer's confirmed fix, never to decide scope on its own.
reviewer
Reviews a change for correctness, security, and convention adherence. Use AFTER a resolver edit and BEFORE merge. Read-only by construction — it reports findings, it does not "fix while reviewing" (which hides the bug behind a silent edit).
codebase-onboarding-engineer
First-pass analyst for unfamiliar codebases — maps tech stack, architecture layers, conventions, hot paths, and known constraints into a concise onboarding brief.
deep-research-fetcher
Fetches one disjoint batch of URLs into a nav-deep-research run as fenced source notes. Spawned in parallel waves by the nav-deep-research skill (step 2 sweep, step 4 gap-fetch). Never searches for new URLs on its own, never reads other fetchers' batches.
scout
Codebase reconnaissance agent. Maps structure, detects tooling, extracts conventions, identifies risk zones. Spawned by orchestrator on first interaction with unfamiliar codebase or when discovery is needed.