Paper Scout
Find and verify real research papers through a keyless index — no login, no API key. Search by topic or method, map a field's citation graph, pull the in-body passages that prove a claim, or scan GitHub for engineering prior art. Grounds every answer in papers the index actually returns — never invents titles or arXiv IDs.
About
Agents hallucinate citations. Paper Scout reads the papers first.
Five moves over one keyless index: `search` (semantic search over abstracts), `similar` (turn one hit into its family — references, citers, or neighbours), `inspect` (canonical metadata for one paper), `read` (the in-body passages that answer a specific question — this is how you verify), and `github` (issues / PRs / READMEs for engineering prior art).
Six use cases: find the paper, map a field or related-work, ground a claim before you assert it (the anti-hallucination move for any writing agent), find prior art, scan what's new, and build a reading list — references are the prerequisites, citers are the frontier.
Honest by construction: stdlib Python, one file, no dependencies. It returns only ids, titles, and numbers the index actually returned — never an invented arXiv id. Passages come back scored, and a low score is treated as unconfirmed, not laundered into a citation. When the index can't answer, `search` degrades out loud to the public arXiv API with a stderr banner; the other moves print the error instead of guessing.
Keyless by default. The research index answers over plain HTTP with no login and no API key, capped per IP per day; an optional free `FIRECRAWL_API_KEY` lifts the ceiling, never gates it. Data backend is Firecrawl's public research index — the skill, method, engine, and voice are Solid State's own.
A daily investor-grade briefing on SpaceX (now public as SPCX, merged with xAI). Live X + web search across the stock, Starlink, and the xAI/Grok AI segment — with Nasdaq/Yahoo analyst targets, SEC filings, and Kalshi prediction-market odds — distilled into a tight, price-moving, sourced brief.
Surface profitable affiliate sub-niches scored by commission rate, SERP gap, trend slope, and content difficulty.
Fair, structured 1v1 comparison between two AI tools. Pricing, features, real-world performance, switching cost, who-it's-for.