agentleFS
hemapriyan-rkhemapriyan-rk/claude-rigor-skills/skills/researcher/SKILL.md

researcher

Skill1 starsChanged 52 days agoMITView on GitHub

Adversarial technical research that tries to break the claim, not confirm it. Covers papers, patents, shipped products, and standards, with a search budget sized to the stakes. Use to check prior art, map the state of the art, or validate a claim before committing to it. Not for one-fact lookups.

42 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use build-premium-website, albert-dm and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • yawningwhaleAISkill

    build-premium-website

    Build a premium, animated marketing website (React + Vite + Tailwind + GSAP) for any industry. Use when the user asks to "build a website", "make a landing page for [business]", "build a marketing site", or wants a high-end animated company website. Adapts color, copy, services, and signature animation to the industry.

    Also in 9 projects that use this

  • yawningwhaleAISkill

    albert-dm

    Reply to DMs and messages the way Albert would. Use when the user asks "how would Albert reply to this", pastes a conversation and wants an Albert-style response, or asks you to draft/write a DM in Albert's voice. Albert sells AI voice agents / automations to local service businesses (Go High Level, lead follow-up, bookings).

    Also in 7 projects that use this

  • huyvpSkill

    cost-reducer

    Reduce cloud, infrastructure, and operational costs while maintaining performance. Use when writing database queries, configuring cloud resources, optimizing bundles, setting up caching, choosing between services, sizing instances, configuring CDN, managing storage, or reviewing code for cost inefficiencies. Covers AWS/GCP/Vercel pricing, database optimization, serverless tuning, image pipelines, observability costs, and FinOps practices.

    Also in 13 projects that use this

  • RoxenRobloxSkill

    youtube-broll-maker

    Generate 10-second B-roll footage of yourself doing an action using the Higgsfield MCP (Seedance 2.0). Upload one photo of yourself, describe the action ("typing on laptop", "walking at night"), pick from 5 cinematography style presets, get an MP4. Use when the user wants B-roll, cutaway footage, "me doing X" video, YouTube B-roll, or any image-to-video of themselves performing an action.

    Also in 4 projects that use this

  • RoxenRobloxSkill

    youtube-popup-graphic

    Generate stunning Adobe-Premiere-Pro-style UI popup motion graphics for YouTube videos using the Higgsfield MCP. The user uploads a style reference image (any UI mockup, dashboard, glass card, design they love), names one skill/feature/item, and the skill produces an MP4 of a single card popping in styled to match the reference. Use when the user wants a "popup graphic", "popup animation", "card popup", "skill popup", "AE/Premiere-style animation", "stunning UI animation for YouTube", or says "make me a popup for <skill/feature/item> styled like <reference>".

    Also in 4 projects that use this

  • xfstudioSkill

    trigger-dev

    Trigger.dev expert for background jobs, AI workflows, and reliable async execution with excellent developer experience and TypeScript-first design. Use when: trigger.dev, trigger dev, background task, ai background job, long running task.

    Also in 12 projects that use this

  • +4 more from yawningwhaleAI

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 38 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before any search, write down:

    line 12
  2. Distinctive:

    Default to Standard unless the user says otherwise.

    line 24
  3. Distinctive:

    Stop early: when you find the kill evidence (report it), or when every relevant layer has been searched and fresh queries only return …

    line 26
  4. Very specific:

    Read references/sources.md from this skill's base directory before the first search.

    line 28
  5. Distinctive:

    Standards and regulation: when the domain has them.

    line 37
  6. Distinctive:

    Don't treat marketing pages or a company's claims about itself as verified fact.

    line 71

“Common” is measured: kept by at least 6 in 10 of the 38 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Researcher

Most research fails because it searches to confirm, not to falsify. This skill produces research that survives a patent examiner, a peer reviewer, or a skeptical cofounder, and spends only as many searches as the decision is worth.

1. Scope and budget

Before any search, write down:

  • The claim being tested, stated so it can be false: "A system that does X using Y under constraint Z doesn't exist," not "is my idea good?"
  • What would kill it: the specific evidence that would make the claim false. If you can't name it, you don't understand the claim yet.
  • The search budget, based on what the answer will decide:
StakesBudgetExample
Quickup to 5 searches, 2 page readsIs this idea worth an afternoon?
Standardup to 15 searches, 6 page readsChoosing an approach for a project
Deepup to 40 searches, 15 page readsPatent filing, paper submission, months of work

Default to Standard unless the user says otherwise. Say which budget you're using.

Stop early when you find the kill evidence (report it), or when every relevant layer has been searched and fresh queries only return sources you've already seen.

Read references/sources.md from this skill's base directory before the first search. It lists where to search for each layer and how to build queries.

2. Search every layer that matters

Skipping a layer is the most common way research gets blindsided. Cover the ones relevant to the claim:

  • Academic: papers and preprints.
  • Patents: required whenever novelty is even part of the question. Patents rarely show up in plain web search.
  • Shipped products and open source: what actually exists and works, as opposed to what was published.
  • Standards and regulation: when the domain has them.

3. Search for the kill

After a favorable first pass, spend part of the budget trying to break the conclusion. Search with the words a competitor or examiner would use, describing the mechanism, not the user's own terms. Different names for the same thing hide the closest prior art from naive searches.

4. Double-source what the conclusion rests on

Find the two or three claims the conclusion depends on. Each needs two independent sources. Mark everything else as single-sourced.

Output

# Research: <topic>
Budget: <Quick / Standard / Deep>, <n> searches used.

## Claim tested
## What would falsify it (stated before searching)

## Closest existing work
| Source (dated) | Type | Mechanism | How close | Gap vs. the claim |

## Strongest counter-evidence
<the best evidence against the claim, or "none found" plus which layers and queries were checked>

## Verdict
<supported / weakened / killed>, in one or two sentences, with confidence and why.

## Open questions
## Sources (dated, grouped by type)

Rules for the write-up:

  • Don't treat marketing pages or a company's claims about itself as verified fact.
  • "No counter-evidence found" must say how hard you looked.
  • Hand the result to researcher-evaluator for a second pass when the stakes are Deep.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.