ClawHub

OkraPDF Public Documents

v1.0.0

Query pre-extracted public documents via OkraPDF MCP — arxiv AI papers, SEC 10-K/10-Q filings, and more. Read, ask questions, extract structured data. No upl...

⭐ 0· 52·0 current·0 all-time
bySteven Tsao@steventsao
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (querying pre-extracted ArXiv and SEC docs) matches the provided assets and instructions: a papers.json manifest and MCP configuration examples. No unrelated credentials, binaries, or install steps are requested.
✓
Instruction Scope
SKILL.md confines runtime instructions to querying documents via MCP endpoints and optional public search APIs (Semantic Scholar, PapersWithCode). It asks the user to add MCP server config to client config files (~/.claude/mcp.json or .cursor/mcp.json) and to provide an OkraPDF API key only for the ArXiv channel — this is proportional to the stated functionality. It does not instruct scanning of unrelated local files or exfiltration.
✓
Install Mechanism
No install spec or downloaded code is present (instruction-only), so nothing is written to disk or executed by the skill beyond guidance to configure a client to call external MCP endpoints.
✓
Credentials
No required environment variables or hidden credentials are declared. The only credential mentioned is an optional OkraPDF API key for the ArXiv channel, which is appropriate for authenticated API usage; SEC filings are described as zero-auth.
✓
Persistence & Privilege
Skill is not always-enabled and is user-invocable. It does not request persistent system privileges or modify other skills' configurations; the only persistent change it recommends is adding MCP entries to the user's agent/client config (which is a normal client-side setup step).
Assessment
This is an instruction-only connector that tells your agent how to use OkraPDF's MCP endpoints to query pre-indexed public documents. Before installing: 1) Confirm you trust api.okrapdf.com / mcp.okrapdf.com because the agent will send queries there; 2) only provide an OkraPDF API key if you accept that queries referencing the ArXiv collection will go to that service; 3) the skill asks you to edit client config files (~/.claude/mcp.json or .cursor/mcp.json) — review those edits before saving; 4) the SEC data is described as public and zero-auth but verify data quality if you rely on it for decisions. Nothing in the skill demands unrelated credentials or elevated system access.

Like a lobster shell, security has layers — review code before you run it.

latestvk977gbkc4sgrjn0spx30rnd4th83xprz

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments