OpenSkillIndex← back to search
claude-code

Autosearch dblp

Skill Attribution

install
source · Clone the upstream repo
git clone https://github.com/0xmariowu/Autosearch
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/0xmariowu/Autosearch "$T" && mkdir -p ~/.claude/skills && cp -r "$T/autosearch/skills/channels/dblp" ~/.claude/skills/0xmariowu-autosearch-dblp && rm -rf "$T"
manifest: autosearch/skills/channels/dblp/SKILL.md
source content

Skill Attribution

Source: self-written for task Plan-0420 W7 F701 + F702.

name: dblp description: Computer science bibliography search — technical papers, proceedings, and journal articles indexed by venue, author, and year. version: 1 languages: [en] methods:

  • id: api_search impl: methods/api_search.py requires: [] rate_limit: {per_min: 30, per_hour: 500} fallback_chain: [api_search] when_to_use: query_languages: [en] query_types: [academic, computer-science, research, technical-paper] avoid_for: [social, multimedia, real-time] quality_hint: typical_yield: medium chinese_native: false layer: leaf domains: [academic] scenarios: [computer-science, author-bibliography] model_tier: Fast experience_digest: experience.md

Overview

dblp
is the focused bibliography channel for computer science papers, conference proceedings, and journal articles. It is useful when the query is clearly CS-oriented and benefits from venue, author, and year metadata even when abstracts are unavailable.

When To Choose It

  • Choose it for English CS paper discovery, author lookup, venue lookup, and bibliography-style queries.
  • Choose it when the planner wants broader CS venue coverage than a single preprint index.
  • Avoid it for social discussion, multimedia content, or breaking news.

How To Search

  • api_search
    queries the DBLP publication API and normalizes hits into evidence with external links when available.
  • Result snippets are synthesized from authors, venue, year, and record type because DBLP does not provide abstracts.

Known Quirks

  • Titles may end with a trailing period in the raw API payload and should be normalized.
  • Some hits only expose a DBLP record page while others provide an external DOI landing page.
  • Coverage is strongest for computer science and adjacent technical literature, not general scholarly search.

Quality Bar

  • Evidence items have non-empty title and url.
  • No crash on empty or malformed API response.
  • Source channel field matches the channel name.