Computer Science bibliography search server for DBLP (Digital Bibliography & Library Project)
Three tools with clear verb-noun naming (search_*) and basic schemas. All tools have descriptions (10 - 100 chars, baseline 194 avg) and input parameters with types. However, descriptions are minimal (e.g., 'Search DBLP venues' lacks context on when to use vs. search_publications). No output schemas documented. Parameters lack constraints (hits 1 - 1000 stated in description but not enforced via schema minima/maxima). No error handling guidance. Missing pagination metadata (total count, next_cursor). Tools are read-only but lack explicit idempotentHint annotations.
Search DBLP authors by name. Returns canonical id, affiliation, ORCID when available.
Search DBLP publications. Match by title / author / year / venue. Default page size 30, max 1000.
Search DBLP venues (conferences + journals).
Output schemas not documented. LLMs cannot infer response structure (fields, types, pagination). Agents cannot plan downstream calls or extract required data.
Parameter constraints (hits: 1 - 1000, first: ≥0) stated in descriptions but not enforced in JSON Schema (missing minimum/maximum). LLMs cannot validate inputs; invalid values reach the API.
Descriptions are minimal (55 - 65 chars). Lack context on when to use each tool (e.g., search_publications vs. search_authors vs. search_venues). No guidance on expected output or prerequisites.
No pagination metadata returned (total count, next_cursor). Large result sets will blow context window. No guidance on result limits or how to fetch next page.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 54 | 2026-07-28+ | v2 |
No tool annotations (idempotentHint, readOnlyHint). All three tools are read-only and idempotent, but LLMs cannot infer this without explicit hints.