Claude Scientific Writer is an AI-assisted research and writing tool that searches literature and produces documents such as scientific papers, reports, posters, grant proposals, and reviews with citations. Researchers and technical writers can use it as a Claude Code plugin, Python package, or command-line tool, with the catalogue entries defining agent workflows for it.
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/K-Dense-AI/claude-scientific-writerWrote 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/commands/k-dense-ai/claude-scientific-writer/scientific-writer-init)<a href="https://agentmods.dev/commands/k-dense-ai/claude-scientific-writer/scientific-writer-init"><img src="https://agentmods.dev/badge/commands/k-dense-ai/claude-scientific-writer/scientific-writer-init/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/commands/k-dense-ai/claude-scientific-writer/scientific-writer-init"><img src="https://agentmods.dev/badge/commands/k-dense-ai/claude-scientific-writer/scientific-writer-init.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 | $0.00026 | $0.01578 |
| Opus 5 | $0.00013 | $0.00789 |
| Sonnet 5 | $0.00005 | $0.00316 |
| Haiku 4.5 | $0.00003 | $0.00158 |
Grade A, and why
scientific-writer-init 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 12d 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 — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scientific Writer Project Setup
When the user runs /claude-scientific-writer:scientific-writer-init, do the following:
⚠️ CRITICAL RULE: NEVER READ THE TEMPLATE FILE
Throughout this entire process, you must NEVER use the read_file tool on a template file. The template files are very long and reading them wastes time and tokens. Only use terminal commands (cp, cat, mv) to handle them.
Instruction Files Installed
Scientific Writer configures two project instruction files so both Claude Code and other agent clients pick it up:
| Project file | Template |
|---|---|
CLAUDE.md |
CLAUDE.scientific-writer.md |
AGENTS.md |
AGENTS.scientific-writer.md |
The two templates hold identical instructions; only the generated header differs. Steps 1 and 3 apply to each of these files independently.
Step 1: Check for Existing Instruction Files
-
Check whether
CLAUDE.mdand/orAGENTS.mdexist in the current working directory. -
For each file that already exists:
- Ask the user whether to:
- a) Back up the existing file as
<FILE>.bakand replace it with the Scientific Writer configuration, or - b) Merge the Scientific Writer settings into the existing file (append to end), or
- c) Skip this file.
- a) Back up the existing file as
- If both files exist, ask once and apply the same choice to both unless the user says otherwise.
- Ask the user whether to:
-
If the user wants to cancel the whole operation, stop here.
-
Wait for user response before proceeding.
Step 2: Locate the Template File Paths
CRITICAL: Do NOT use the read_file tool. Do NOT read the template contents. Only locate the file paths.
Find the paths to the Scientific Writer templates. Use one of these methods:
- Use glob_file_search to find
CLAUDE.scientific-writer.mdandAGENTS.scientific-writer.mdin the templates directory - Use list_dir to check if the files exist in known locations
- Directly try these paths (in order):
~/.claude/plugins/*/claude-scientific-writer/templates/(installed plugin)<path-to-claude-scientific-writer-checkout>/templates/(local development)
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.
- 12d ago First seen · 160 lines · 26 tokens per session scan A 95ce829bcb48
scientific-writer-init is a command published in the GitHub repository K-Dense-AI/claude-scientific-writer (2,324 stars, last pushed 23d ago), licensed MIT. It adds 26 tokens to every session and 1,578 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 2026-08-30.
Other commands, from other repositories
whats-next
Analyze the current conversation and create a handoff document for continuing this work in a fresh context.
add-to-todos
Add todo item to TO-DOS.md with context from conversation.
check-todos
List outstanding todos and select one to work on.
create-prompt
Create a new prompt that another Claude can execute.
add-skill
Scaffold a new skill with validation, plugin generation, and README update.
debug
Apply expert debugging methodology to investigate a specific issue.