Skip to content
SkillModel-led123 tokens in use

Research

Delegates reading to a background agent that investigates primary sources and saves cited findings as a Markdown file in your repo.

328,754 stars on the repository MIT Licensemattpocock/skills · skills/engineering/research
claude plugin install mattpocock-skills@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Research costs

49tokens idle
123tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts and code blocks. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter12/15
  • Structure12/25
  • Examples0/15
  • Output5/15
  • Size15/15
  • Resources0/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community56
  • Maturity32
  • Momentum40

Research starts a background agent with a fresh context window, so you keep working while it reads. The agent investigates a question against primary sources such as official docs, source code, specs and first-party APIs, follows each claim back to the source that owns it, and writes the findings to one Markdown file.

The file lands where the repo already keeps such notes, or somewhere sensible that the agent names.

What You Get

  • A single Markdown file of findings
  • A citation for each claim that points to its primary source
  • A save location that matches the repo’s existing notes convention
  • A background run that leaves your session free

Common Use Cases

  • Gathering the facts of a third-party API before designing against it
  • Checking from a library’s source code what a given option does
  • Preparing cited background for a grilling session

What Makes It Unique

It is three steps long and puts its weight on source quality: primary sources only, with each claim traced to the party that owns it. The agent runs in the background, and the output is a file in the repo that later skills can read.

Under The Hood

SKILL.md is a short Markdown instruction with no scripts. The model starts a background agent that reads web pages and source code, then writes one Markdown file in your repository. Which sources it reads depends on your question.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers