sciverse academic retrieval
PassAudited by VirusTotal on May 12, 2026.
Overview
Type: OpenClaw Skill Name: academic-retrieval Version: 0.1.6 The skill bundle provides legitimate tools for academic paper retrieval via the SciVerse API. The code is clean, uses native Node.js fetch without external dependencies, and includes a security validation check in `scripts/_common.mjs` to ensure the API token is only sent to authorized `*.sciverse.space` domains, preventing credential leakage.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Search queries and content requests will be sent to SciVerse using the configured API token.
The skill authenticates to the provider API using an environment token. This is consistent with its SciVerse retrieval purpose, but users should understand that tool calls are made under their SciVerse API identity.
const TOKEN = process.env.SCIVERSE_API_TOKEN; ... authorization: `Bearer ${TOKEN}`Use a SciVerse token intended for this integration, keep it private, and prefer the default HTTPS endpoint unless you knowingly configure another approved SciVerse endpoint.
Users have less registry-level information for independently verifying the publisher/source of the skill.
The registry metadata does not provide a source repository or homepage, which limits provenance verification even though the included code has no external dependencies.
Source: unknown Homepage: none
Verify that this is the intended SciVerse package before installing, especially because it uses an API token.
A retrieved paper excerpt could affect generated answers if the agent treats it as authoritative beyond its citation value.
The skill intentionally returns external text chunks for use in RAG. Retrieved content can influence an agent's answer and should be handled as untrusted source material.
semantic_search ... Natural-language semantic search returning relevant paper chunks for RAG-style answering.
Use retrieved chunks as evidence to cite and verify, not as instructions that override the user's goal or safety constraints.