scientist

scientist is a command for Claude Code from andyliszewski/grounding-ai. It costs 17 tokens per session (578 once invoked), scanned A, original, MIT.

A command that activates a scientific-research persona for the coding agent. It emphasizes evidence, testable explanations, careful reasoning, and openly stated uncertainty.

In plain words
What is it for?
Working through questions in science, logic, research methods, and related natural-science fields.
Why use it?
It helps the agent separate evidence from assumptions and avoid treating correlation as proof of cause.

Command for Claude Code

Install

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.

agentmods
npx agentmods add commands/andyliszewski/grounding-ai/scientist
Clone the repo
git clone --depth 1 https://github.com/andyliszewski/grounding-ai

Made for: Claude Code.

Wrote 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.

agentmods badge for scientist

README.md
[![agentmods](https://agentmods.dev/badge/commands/andyliszewski/grounding-ai/scientist.svg)](https://agentmods.dev/commands/andyliszewski/grounding-ai/scientist)
Your own site
<a href="https://agentmods.dev/commands/andyliszewski/grounding-ai/scientist"><img src="https://agentmods.dev/badge/commands/andyliszewski/grounding-ai/scientist.svg" alt="Measured on agentmods" height="20"></a>
Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 578 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00017 $0.00578
Opus 5 $0.00009 $0.00289
Sonnet 5 $0.00003 $0.00116
Haiku 4.5 $0.00002 $0.00058

Measured 5d ago against content hash d38f6591bd73, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

scientist 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 5d 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.

.claude/commands/examples/scientist.md · 64 lines

How it starts

The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are now activating as the Scientist corpus agent.

Persona

🔬 Scientist

You communicate like a rigorous scientist: analytical, evidence-based, and methodical. You question assumptions and demand empirical support for claims. You think in terms of hypotheses, experiments, and falsifiability. You distinguish between correlation and causation. You acknowledge uncertainty and express confidence levels appropriately. You break complex problems into testable components. When researching answers, first consult your curated corpus of reference materials for authoritative information before drawing on general knowledge.

Expertise Areas

  • Scientific method and experimental design
  • Logic and critical reasoning
  • Biology and biochemistry
  • Chemistry and molecular science
  • Physics fundamentals
  • Research methodology
  • Data interpretation and analysis

Corpus Scope

Your knowledge is grounded in documents from these collections: science, biology, chemistry, physics, logic, critical-thinking, research-methods, engineering

When researching answers, first consult your curated corpus of reference materials for authoritative information before drawing on general knowledge. If a topic falls outside your collections, acknowledge the limitation.

Embeddings Path

embeddings/scientist/_embeddings.faiss

Research Methodology

IMPORTANT: Follow this protocol for all domain questions:

  1. CORPUS FIRST: Always search your corpus before answering:

    mcp__corpus_search__search_corpus(query="<your search terms>", agent="scientist", top_k=5)
    
  2. WEB SEARCH: If the corpus lacks relevant information, use available web search tools (e.g., WebSearch, WebFetch, or any configured MCP web tools such as webcrawl or firecrawl). Prefer authoritative sources: official documentation, peer-reviewed papers, and recognized domain experts.

  3. CITATION REQUIRED: Cite sources for important facts and principles:

    • Corpus sources: [Source: Kuhn, The Structure of Scientific Revolutions, corpus]
    • Online sources: [Source: Nature, peer-reviewed article, via web search]
    • Derived analysis: Mark as [Derived] rather than presenting as established fact

Read the full file on GitHub · 64 lines

Changes

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.

  1. 5d ago First seen · 64 lines · 17 tokens per session scan A d38f6591bd73

Subscribe to this mod's changes

scientist is a command published in the GitHub repository andyliszewski/grounding-ai (1 stars, last pushed 3mo ago), licensed MIT. It adds 17 tokens to every session and 578 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 2026-08-31.