Agent skill

Save Research Notebook

by napjon in napjon/krisk

Convert a completed data-analysis conversation into evidence-backed, reproducible living research through the Krisk MCP server.

BSD-3-ClauseAuto-check passedData & Analytics

Install Save Research Notebook

skills CLI
$ npx skills add napjon/krisk --skill save-research-notebook -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install napjon/krisk save-research-notebook --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/napjon/krisk.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/save-research-notebook .claude/skills/save-research-notebook && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
save-research-notebook
GitHub stars
118
Token cost
~702 tokens
SKILL.md length
339 words
Files
4 (incl. references)
Skills in repo
1
Repo updated
First seen
Licence
BSD-3-Clause

At a glance

Convert a completed data-analysis conversation into evidence-backed, reproducible living research through the Krisk MCP server.

  • Works in 9 steps: Confirm that the user wants to save the… → Identify the research question,… → Reuse only query IDs and chart IDs… → …
  • A user asks to save an investigation
  • SKILL.md covers Workflow, Evidence Rules and Failure Handling
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Save Research Notebook is an agent skill from napjon/krisk. Convert a completed data-analysis conversation into evidence-backed, reproducible living research through the Krisk MCP server. Use when a user asks to save an investigation, formalize its research methodology, preserve point-in-time charts, create executable conclusion checks, or export the work as a Jupyter notebook or static/live HTML report.

Its SKILL.md is about 700 tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `agents/openai.yaml`, `references/claim-types.md` and `references/research-schema.md`).

It sits in Data & Analytics, covering Source-grounded notebooks, Jupyter notebooks and Experimental design. It works with Jupyter, Model Context Protocol and Python. The repository describes itself as: Statistical Interactive Visualization with pandas+Jupyter integration on top of Echarts. The licence is BSD-3-Clause.

When your agent uses it

  • A user asks to save an investigation
  • Formalize its research methodology
  • Preserve point-in-time charts
  • Create executable conclusion checks

Example prompts

  • “/save-research-notebook”

Workflow steps

9 steps, taken from the first numbered list in SKILL.md.

  1. Confirm that the user wants to save the work. Do not persist an ordinary in-progress conversation.
  2. Identify the research question, hypothesis if any, ordered methodology, findings, conclusion, and limitations from the conversation.
  3. Reuse only query IDs and chart IDs returned by Krisk tools. If a necessary claim lacks evidence, run a new read-only query and show its…
  4. Distinguish observations from inferences. State the capture time or data cutoff for every point-in-time finding.
  5. Read references/research-schema.md before constructing save_research input.
  6. Read references/claim-types.md when a conclusion can be checked against future data. Keep subjective or causal conclusions narrative-only.
  7. Show the proposed conclusion, limitations, and executable claims to the user. Resolve material corrections before saving.
  8. Call save_research once. Then call export_research when the user requested a notebook or HTML export.
  9. Return the research ID, artifact locations, snapshot timestamp, and the command or link used to refresh live evidence.

What it can do on your machine

Read from SKILL.md and the folder at commit 73b0409. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Save Research Notebook loads about 702 tokens when it runs, and up to ~1.3k if it reads all its reference files. Until then it costs about 93 tokens; SKILL.md has 339 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~93
When it runs · the whole SKILL.md, loaded when a task matches
~702
With references · SKILL.md plus every file in references/, read only if the agent opens them
~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from napjon/krisk at commit 73b0409, republished under its BSD-3-Clause licence (© napjon). 339 words, ~702 tokens.

Download SKILL.mdSave it as .claude/skills/save-research-notebook/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
save-research-notebook
description
Convert a completed data-analysis conversation into evidence-backed, reproducible living research through the Krisk MCP server. Use when a user asks to save an investigation, formalize its research methodology, preserve point-in-time charts, create executable conclusion checks, or export the work as a Jupyter notebook or static/live HTML report.

Save Research Notebook

Turn an explicitly concluded investigation into a structured Krisk research record and export. Preserve observed evidence; never invent or silently recompute facts.

Workflow

  1. Confirm that the user wants to save the work. Do not persist an ordinary in-progress conversation.
  2. Identify the research question, hypothesis if any, ordered methodology, findings, conclusion, and limitations from the conversation.
  3. Reuse only query IDs and chart IDs returned by Krisk tools. If a necessary claim lacks evidence, run a new read-only query and show its result before saving.
  4. Distinguish observations from inferences. State the capture time or data cutoff for every point-in-time finding.
  5. Read references/research-schema.md before constructing save_research input.
  6. Read references/claim-types.md when a conclusion can be checked against future data. Keep subjective or causal conclusions narrative-only.
  7. Show the proposed conclusion, limitations, and executable claims to the user. Resolve material corrections before saving.
  8. Call save_research once. Then call export_research when the user requested a notebook or HTML export.
  9. Return the research ID, artifact locations, snapshot timestamp, and the command or link used to refresh live evidence.

Evidence Rules

  • Treat MCP query output as observation; treat interpretation as inference.
  • Include query and chart references in the structured record, not credentials or connection URIs.
  • Preserve the original snapshot and conclusion even when current data differs.
  • Use a typed claim only when one bounded query returns the scalar values needed to evaluate it.
  • Record missing data, sampling choices, filters, failed queries, and alternative explanations as limitations.
  • If Krisk MCP is unavailable, stop before claiming the research was saved and provide the local setup command.

Failure Handling

  • If a referenced query or chart no longer exists, recreate the evidence with the user’s knowledge; never substitute an unrelated record.
  • If a claim query returns more than one row, revise it to a one-row metric query before saving.
  • If export fails, retain the saved research ID and report the export failure separately.
  • If current data cannot be fetched, keep the static snapshot usable and mark live validity as unknown.

© napjon, BSD-3-Clause. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (references) in skills/save-research-notebook of napjon/krisk.

  • SKILL.md
  • agents/openai.yaml
  • references/claim-types.md
  • references/research-schema.md

Open the folder on GitHubat commit 73b0409

Compare with similar skills

Save Research Notebook next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Save Research Notebook compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Save Research Notebook this skillnapjon/krisk118—~702Automated safety check: PassBSD-3-Clause
Jupyter Notebook BuilderKilo-Org/kilo-marketplace190—~1.3kAutomated safety check: PassMIT
Analytics Data AnalysisMindrally/skills268—~1.6kAutomated safety check: PassApache-2.0
Python Executorcortega26/chile-hub1132 repos~1.5kAutomated safety check: PassMIT
Raccoon DataanalysisSenseTime-Copilot/raccoon-dataanalysis-skill137—~1.9kAutomated safety check: PassNone
Noteworthysihooleebd/noteworthy120—~2kAutomated safety check: PassMIT

Similar skills

  • Jupyter Notebook Builder

    Kilo-Org/kilo-marketplace

    Creates, inspects, edits and runs Jupyter notebooks, scaffolding experiment or tutorial notebooks from templates and preferring a Jupyter MCP server over raw JSON edits.

    190 GitHub stars~1.3k tokensUpdated 10 days ago
    Data & AnalyticsAuto-check passed
  • Analytics Data Analysis

    Mindrally/skills

    Best practices for analytics, data analysis, and visualization using Python, pandas, matplotlib, seaborn, and Jupyter notebooks.

    268 GitHub stars~1.6k tokensUpdated 1 mo ago
    Data & AnalyticsAuto-check passed
  • Python Executor

    cortega26/chile-hub

    Execute Python code in a safe sandboxed environment via [inference.sh](https://inference.sh).

    113 GitHub starsUsed in 2 repos~1.5k tokens
    Data & AnalyticsAuto-check passed
  • Raccoon Dataanalysis

    SenseTime-Copilot/raccoon-dataanalysis-skill

    Raccoon (小浣熊) Data Analysis - Remote code interpreter and data visualization service powered by SenseTime.

    137 GitHub stars~1.9k tokensUpdated 6 mo ago
    Data & AnalyticsAuto-check passed
  • Noteworthy

    sihooleebd/noteworthy

    Write and edit Noteworthy documents — a Typst framework for textbooks, with blocks (definition/theorem/example/solution), 2D and 3D canvases, geometry, plotting, and live Yjs collaboration.

    120 GitHub stars~2k tokensUpdated 9 days ago
    Data & AnalyticsAuto-check passed
  • Release Evidence Workflow

    Ali-Marandi/ClimateDataAnalyzer

    Build an auditable release-evidence workflow for a desktop or packaged application.

    107 GitHub stars~1.6k tokensUpdated 1 mo ago
    Data & AnalyticsAuto-check passed

Questions about Save Research Notebook

What does Save Research Notebook do?

Convert a completed data-analysis conversation into evidence-backed, reproducible living research through the Krisk MCP server. Save Research Notebook is an agent skill from napjon/krisk. Convert a completed data-analysis conversation into evidence-backed, reproducible living research through the Krisk MCP server.

When should I use Save Research Notebook?

Save Research Notebook fits situations like: A user asks to save an investigation; formalize its research methodology; preserve point-in-time charts; create executable conclusion checks.

How do I install Save Research Notebook in Claude Code?

Run `npx skills add napjon/krisk --skill save-research-notebook -a claude-code`. Or copy the skill folder (skills/save-research-notebook in napjon/krisk) into .claude/skills/save-research-notebook in your project. Claude Code loads it when a task matches its description.

How do I install Save Research Notebook in Codex?

Run `npx skills add napjon/krisk --skill save-research-notebook -a codex`. Or copy the skill folder (skills/save-research-notebook in napjon/krisk) into .agents/skills/save-research-notebook in your project. Codex loads it when a task matches its description.

Can I use Save Research Notebook in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add napjon/krisk --skill save-research-notebook -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/save-research-notebook, .gemini/skills/save-research-notebook, .github/skills/save-research-notebook and .opencode/skills/save-research-notebook in your project.

What does Save Research Notebook need to run?

SKILL.md names no scripts, command-line tools or credentials: Save Research Notebook is instructions for the agent only.

Does Save Research Notebook access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Save Research Notebook safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Save Research Notebook use?

Save Research Notebook is published under the BSD-3-Clause licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Save Research Notebook use?

About 702 tokens (SKILL.md is roughly 2.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 580 tokens, read only when the agent opens those files.

What are the alternatives to Save Research Notebook?

Skills that share tags, products or a category with Save Research Notebook: Jupyter Notebook Builder (Kilo-Org/kilo-marketplace, 190 stars), Analytics Data Analysis (Mindrally/skills, 268 stars), Python Executor (cortega26/chile-hub, 113 stars) and Raccoon Dataanalysis (SenseTime-Copilot/raccoon-dataanalysis-skill, 137 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Save Research Notebook?

napjon (a GitHub user) maintains it in napjon/krisk, which has 118 GitHub stars. The repository was last updated on October 1, 2026.

Source: napjon/krisk on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.