Skip to content
Get packs

Social Evidence Research

Build social media research reports where every claim traces back to inspected source material

Source Download skill
Repository
wshobson/agents
Stars
42,095
Licence
MIT
Files
4 files

What it does

This skill teaches an AI agent to research how a product, event, or topic is discussed on social platforms, using public material or exports the user is authorized to share. The agent keeps a record of what it actually inspected, who said what, which reposts are duplicates, and what it could not access, so every claim in the final report points to specific evidence. It also works as an audit checklist for reviewing existing social research reports for common errors like overreaching conclusions or treating empty search results as proof of absence.

Great for

  • Marketers or analysts who need a cited, checkable summary of social discussion about a product or topic
  • Researchers comparing discussion across platforms using exports they were given
  • Reviewers who want to check an existing social media report for unsupported claims

Skip it if

  • Real-time monitoring or large-scale statistical claims about audience opinion, since the method limits conclusions to the inspected sample and a convenience sample cannot show prevalence
  • Work that requires logging into accounts, scraping past paywalls or rate limits, or posting or messaging, which the skill explicitly prohibits

Try asking

  • “Research how people discuss our new headphone launch on public social posts and build a cited report with a claim table”
  • “Here is a CSV export of our brand mentions; summarize the claims and flag any double-counted reposts”
  • “Audit this social media research report and flag claims that rely only on search snippets”
  • “Compare how two product releases are discussed in the supplied captures, listing access gaps for anything unreachable”
Step-by-step guide: how to use Social Evidence ResearchInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 4 original files, the MIT licence and an attribution note.

    social-evidence-research.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip social-evidence-research.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip social-evidence-research.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Set the research question, platforms, date range, and sample limit before collecting so the protocol is recorded up front
  • Provide exports or captures when possible; the skill treats links you cannot open as discovery-only and supports no conclusions about their content
  • Expect partial reports: the skill is designed to state what is missing rather than fill gaps with guesses

What's inside

  • references/acceptance-examples.md
  • references/evidence-record.md
  • references/report-template.md
  • SKILL.md

Redistributed unmodified from wshobson/agents under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.