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 agentmods add skills/anthropics/knowledge-work-plugins/scientific-problem-selectionnpx skills add anthropics/knowledge-work-plugins --skill scientific-problem-selectiongit clone --depth 1 https://github.com/anthropics/knowledge-work-pluginsWhat 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 | $0.00118 | $0.02552 |
| Opus 5 | $0.00059 | $0.01276 |
| Sonnet 5 | $0.00024 | $0.00510 |
| Haiku 4.5 | $0.00012 | $0.00255 |
Grade A, and why
scientific-problem-selection 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 2d 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.
Copies of this mod
3 near-identical copies found in the catalogue:
- scientific-problem-selection — 100% identical, 0 lines differ
- scientific-problem-selection — 100% identical, 0 lines differ
- scientific-problem-selection — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 270 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scientific Problem Selection Skills
A conversational framework for systematic scientific problem selection based on Fischbach & Walsh's "Problem choice and decision trees in science and engineering" (Cell, 2024).
Getting Started
Present users with three entry points:
1) Pitch an idea for a new project — to work it up together
2) Share a problem in a current project — to troubleshoot together
3) Ask a strategic question — to navigate the decision tree together
This conversational entry meets scientists where they are and establishes a collaborative tone.
Option 1: Pitch an Idea
Initial Prompt
Ask: "Tell me the short version of your idea (1-2 sentences)."
Response Approach
After the user shares their idea, return a quick summary (no more than one paragraph) demonstrating understanding. Note the general area of research and rephrase the idea in a way that highlights its kernel—showing alignment and readiness to dive into details.
Follow-up Prompt
Then ask for more detail: "Now give me a bit more detail. You might include, however briefly or even say where you are unsure:
- What exactly you want to do
- How you currently plan to do it
- If it works, why will it be a big deal
- What you think are the major risks"
Workflow
From there, guide the user through the early stages of problem selection and evaluation:
- Skill 1: Intuition Pumps - Refine and strengthen the idea
- Skill 2: Risk Assessment - Identify and manage project risks
- Skill 3: Optimization Function - Define success metrics
- Skill 4: Parameter Strategy - Determine what to fix vs. keep flexible
See references/01-intuition-pumps.md, references/02-risk-assessment.md, references/03-optimization-function.md, and references/04-parameter-strategy.md for detailed guidance.
Option 2: Troubleshoot a Problem
Initial Prompt
Ask: "Tell me a short version of your problem (1-2 sentences or whatever is easy)."
Response Approach
After the user shares their problem, return a quick summary (no more than one paragraph) demonstrating understanding. Note the context of the project where the problem occurred and rephrase the problem—highlighting its core essence—so the user knows the situation is understood. Also raise additional questions that seem important to discuss.
What ships with it
10 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.
- LICENSE.txt 10 KB
- references/01-intuition-pumps.md 12 KB
- references/02-risk-assessment.md 13 KB
- references/03-optimization-function.md 18 KB
- references/04-parameter-strategy.md 14 KB
- references/05-decision-tree.md 3.2 KB
- references/06-adversity-planning.md 4.6 KB
- references/07-problem-inversion.md 5.5 KB
- references/08-integration-synthesis.md 5.3 KB
- references/09-meta-framework.md 16 KB
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.
- 2d ago First seen · 270 lines · 118 tokens per session scan A 9ec99e3e5b2f
scientific-problem-selection is a skill published in the GitHub repository anthropics/knowledge-work-plugins (23,791 stars, last pushed today), licensed Apache-2.0. It adds 118 tokens to every session and 2,552 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…