How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

recoup-research-the-web - Skill

Name: recoup-research-the-web
Author: recoupable

Imported from recoupable/skills (skills/recoup-research-the-web/SKILL.md). Install upstream with npx skills add recoupable/skills --skill recoup-research-the-web. Copyright stays with the author.

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: on insufficient credit or entitlement, preserve the task and explain the unavailable feature. Do not surface billing/checkout links, initiate purchases, rely on automatic credit purchases or retry silently.

References

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

Install recoup-research-the-web and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "ocm": "1", "id": "recoupable-skills-recoup-research-the-web", "kind": "skill", "name": "recoup-research-the-web", "description": "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.", "publisher": "recoupable", "version": "1.0.0", "capabilities": { "domains": [ "coding", "finance" ], "tags": [ "skill-md", "skills-sh" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "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." ], "primary": false, "metadata": { "source": { "provider": "skills.sh", "repository": "https://github.com/recoupable/skills", "path": "skills/recoup-research-the-web/SKILL.md", "ref": "HEAD", "url": "https://github.com/recoupable/skills/blob/HEAD/skills/recoup-research-the-web/SKILL.md", "key": "recoupable/skills/skills/recoup-research-the-web/SKILL.md" } }, "instructions": "# Recoup Research — The Web\n\nGeneral open-web research and the graceful-degradation fallback for the other\nresearch skills. Researches any entity, not just artists.\n\n```bash\nexport RECOUP_API=\"https://api.recoupable.dev/api\" # auth header: x-api-key: $RECOUP_API_KEY\n```\n\n## Tools\n\n- `POST /research/web` — quick ranked results (seconds, cheap).\n- `POST /research/deep` — cited narrative (2+ min, expensive).\n- `POST /research/extract {urls, objective}` — read specific URLs.\n- `POST /research/enrich {input, schema}` — structured facts about ANY entity\n (label, manager, venue, brand, person); sc", "cost": { "context_tokens": 327 } }

recoup-research-the-web

About

Recoup Research — The Web

Tools

Guardrails

References

Install recoup-research-the-web and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews