Catalog
mattpocock/research

mattpocock

research

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

NewUpdated Oct 30, 2026

Spin up a background agent to do the research, so you keep working while it reads. Start it with a fresh context window.

Its job:

  1. Investigate the question against primary sources (official docs, source code, specs, first-party APIs), not a secondary write-up of them. Follow every claim back to the source that owns it.
  2. Write the findings to a single Markdown file, citing each claim's source.
  3. Save it where the repo already keeps such notes; match the existing convention, and if there is none, put it somewhere sensible and say where.
Files2
2 files · 1.1 KB

Select a file to preview

Overall Score

58/100

Grade

C

Adequate

Grades are signals, not a certification. Always review a skill yourself before use.

Safety

72

Quality

48

Clarity

62

Completeness

45

Summary

This skill delegates research tasks to a background agent, directing it to investigate questions against primary sources (official docs, source code, APIs) and write findings to a Markdown file. The skill provides minimal but focused guidance on sourcing methodology and output conventions, emphasizing primary source verification over secondary summaries.

Detected Capabilities

background agent spawnfile write (markdown)web/document retrievaldocumentation reading

Trigger Keywords

Phrases that agents use to match this skill to user intent.

research topic backgroundgather documentation factsinvestigate api specsverify from primary sourcescompile research notes

Risk Signals

INFO

Background agent spawning with open-ended research scope

SKILL.md body
WARNING

No constraints on which sources the background agent may access

SKILL.md body
INFO

File write path specified only as 'somewhere sensible' with convention guidance

SKILL.md body

Use Cases

  • Research API documentation to understand current functionality
  • Gather technical specifications from official source materials
  • Investigate best practices from primary sources and standards
  • Compile findings on a topic with source citations for team reference
  • Delegate reading-heavy research tasks while maintaining focus on active work

Quality Notes

  • Positive: Explicit emphasis on primary sources and source verification is a strong quality signal — avoids secondary interpretation bias
  • Positive: Clear delegation model (spin up background agent, keep main agent working) is well-motivated and practical
  • Positive: Output format is well-defined (single Markdown file with citations)
  • Negative: Instruction is extremely brief — only 3 bullet points with minimal detail on research methodology
  • Negative: No guidance on handling ambiguous, contradictory, or missing sources
  • Negative: No examples of acceptable sources or output format shown
  • Negative: No clear success criteria — what makes research 'complete'?
  • Negative: Scope of 'question' is unbounded — no guidance on complexity limits or research depth expectations
  • Negative: No mention of handling time limits, rate limits, or access restrictions from primary sources
  • Negative: 'Somewhere sensible' is vague — needs better guidance on file path conventions
Model: claude-haiku-4-5-20251001Analyzed: Oct 30, 2026

Reviews

Add this skill to your library to leave a review.

No reviews yet

Be the first to share your experience.

Version History

  1. v1.2

    Content updated

    ✦ AISpecifies that background research agent must start with fresh context window

    2026-10-30

    LATEST
  2. v1.1

    Content updated

    ✦ AIAdds agents/openai.yaml configuration file and clarifies primary-source guidance with parenthetical phrasing.

    2026-09-09

    View This Version
  3. v1.0

    2026-07-03

    View This VersionInitial version

Use mattpocock/research in your dev environment

Command Palette

Search for a command to run...