Recoup research the web

Skill recoupable/skills/skills/recoup-research-the-web

Research anything on the open web — quick ranked search, deep cited narrative, read specific URLs, and enrich any entity (label, venue, brand, person) into structured facts. Use for "search the web", "deep research [topic]", "what does this URL say", "enrich [company/label/person]", or as the fallback when structured artist data is thin. Not artist-only. For metrics on a known artist use recoup-research-artist-overview.From its SKILL.md

Install

npx -y skills add recoupable/skills --skill recoup-research-the-web

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.6 KB, as published. Nobody here has run it

Recoup Research — The Web

General open-web research and the graceful-degradation fallback for the other research skills. Researches any entity, not just artists.

export RECOUP_API="https://api.recoupable.dev/api"   # auth header: x-api-key: $RECOUP_API_KEY

Tools

POST /research/web — quick ranked results (seconds, cheap).
POST /research/deep — cited narrative (2+ min, expensive).
POST /research/extract {urls, objective} — read specific URLs.
POST /research/enrich {input, schema} — structured facts about ANY entity (label, manager, venue, brand, person); schema must include "type":"object".

Verify citations (every material claim cited; the source actually supports it; drop dead URLs) and distill to a sourced one-page dossier (fact vs inference vs open question) — not a link dump.

Guardrails

Verify citations; never present an unsupported claim as fact.
Latency: /enrich 60–90s, /deep 2+ min — set timeouts ≥3 min.
Credits: surface checkoutUrl on insufficient_credits.

References

references/workflows.md — web/deep research patterns + entity enrichment.