agentleFS
darrenCWJdarrenCWJ/agent-skills-scaffold/.claude/skills/research/SKILL.md

research

Skill0 starsChanged 3 months agoMITView on GitHub

Multi-source research with source ranking, claim verification, and honest uncertainty. Use whenever gathering facts to answer a question, comparing options, investigating a topic, or collecting material for a report/deck/analysis — any time more than one source will be consulted or a factual deliverable depends on the findings.

189 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use Problem-Solving Techniques, daily-arxiv 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.
  • truongtv22Skill

    Problem-Solving Techniques

    Apply systematic problem-solving techniques for complexity spirals (simplification cascades), innovation blocks (collision-zone thinking), recurring patterns (meta-pattern recognition), assumption constraints (inversion exercise), scale uncertainty (scale game), and dispatch when stuck. Techniques derived from Microsoft Amplifier project patterns adapted for immediate application.

    Also in 6 projects that use this

  • LambenthanSkill

    daily-arxiv

    每日从 arXiv 拉取新论文,过滤相关性,auto-ingest 高优先级论文,检测 SOTA 更新

    Also in 5 projects that use this

  • LambenthanSkill

    exp-eval

    实验判决门:Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

    Also in 5 projects that use this

  • liufeng1976Skill

    setup-ts-deep-modules

    Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.

    Also in 49 projects that use this

  • parcadeiSkill

    ast-grep-find

    AST-based code search and refactoring via ast-grep MCP

    Also in 4 projects that use this

  • SireJeffSkill

    context-optimize

    Context Optimization - Orchestrate RPI workflow TODO generation with interactive scoping

    Also in 4 projects that use this

  • +5 more from Lambenthan

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 170 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Hunt disconfirmation: Before concluding, search explicitly for the opposite claim ("X is not true", "X criticism", "X failed").

    line 16
  2. Distinctive:

    Every non-obvious factual statement carries its source.

    line 22

“Common” is measured: kept by at least 6 in 10 of the 170 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

Research

Research quality = quality of sources × independence of confirmation × honesty about gaps. Volume of text is not a quality metric.

Procedure

  1. Decompose the question. Break into sub-questions, each answerable independently. Note for each: what would a GOOD answer look like (a number, a comparison, a mechanism)?
  2. Plan the search, then fan out. Independent sub-questions with meaningful search depth → parallel subagents (fan-out skill), each briefed with its sub-question, required output format, and source-quality bar. Quick lookups → do inline.
  3. Rank sources as you collect. Tier 1: primary (original papers, official docs, filings, raw data, the actual code). Tier 2: reputable secondary (major outlets, established references, expert analyses). Tier 3: everything else (forums, blogs, SEO content) — leads only, never load-bearing. A Tier 3 claim gets promoted only by finding its Tier 1/2 origin.
  4. Verify load-bearing claims. Any claim the conclusion depends on needs either a Tier 1 source or two INDEPENDENT Tier 2 sources — independent means not citing the same upstream origin; ten articles echoing one press release are one source.
  5. Hunt disconfirmation. Before concluding, search explicitly for the opposite claim ("X is not true", "X criticism", "X failed"). Confirmation-only research is advocacy.
  6. Date-check everything. Note each key fact's as-of date. For anything that changes (prices, versions, laws, leadership, benchmarks), prefer the newest source and state the as-of date in the output.

Synthesis rules

  • Separate FINDINGS (sourced), INFERENCE (your reasoning on findings), and ASSUMPTION (unverified) — visibly, in the deliverable.
  • Every non-obvious factual statement carries its source. Uncited facts in the final output are defects (critique skill treats them as majors).
  • State confidence honestly: "confirmed by primary source" ≠ "single blog post says". If sources conflict, present the conflict — do not silently pick a side.
  • Record what was NOT found: searches that came up empty are findings too, and they save the next researcher the same dead end.

Failure modes to catch in critique

Citation laundering (Tier 3 dressed as fact), stale data presented as current, survivorship of the first plausible answer, and answering a different (easier) question than the one asked.

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.