← Back to Skills Marketplace
347
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install semantic-scholar
Description
Search, retrieve, and organize scholarly metadata with the Semantic Scholar APIs. Use when Codex needs to find papers or authors, build paper sets from compl...
Usage Guidance
This skill appears to do what it says: query Semantic Scholar and save results. Before installing/running: (1) Review and install Python dependencies (requests; pandas only if you need CSV export). The skill has no automated install step. (2) If you want higher rate limits or repeated/bulk jobs, set SEMANTIC_SCHOLAR_API_KEY in your environment; the scripts look for x-api-key but will run without it (with lower limits). (3) The provided smoke-test expects an external 'uv' command (not documented in registry metadata); you don't need to run the smoke-test if 'uv' is unavailable. (4) The scripts perform network requests to api.semanticscholar.org and write JSONL/CSV files to disk—inspect outputs and running directory before sharing them. (5) If you plan to use the Datasets API or large bulk downloads, confirm storage/bandwidth expectations first. Overall: coherent and consistent with the stated purpose, but install/runtime dependencies should be manually verified before execution.
Capability Analysis
Type: OpenClaw Skill
Name: semantic-scholar
Version: 1.0.0
The skill bundle provides a comprehensive and well-structured interface for the Semantic Scholar API, including scripts for bulk search, batch metadata retrieval, and paper recommendations. The Python scripts (e.g., semantic_scholar_bulk_search.py, paper_batch_fetch.py) implement robust error handling with exponential backoff for rate limits and use standard environment variables for API key management. No evidence of data exfiltration, malicious execution, or prompt injection was found; the instructions in SKILL.md and the references are strictly focused on guiding the agent through legitimate API workflows.
Capability Assessment
Purpose & Capability
Name/description match the included scripts and references: Graph API, Recommendations API, Datasets API workflows are implemented or documented. All request URLs point to api.semanticscholar.org and the scripts implement search, batch fetch, recommendations, and dataset guidance—functions consistent with the description.
Instruction Scope
SKILL.md and the scripts limit operations to calling Semantic Scholar endpoints, writing JSONL/CSV outputs, and handling retries/pagination. The instructions do not ask the agent to read unrelated host files or exfiltrate secrets. The scripts preserve raw output before flattening as recommended.
Install Mechanism
There is no registered install spec (skill is effectively delivered as code). The Python scripts declare typical Python deps (requests, optional pandas) but the registry metadata doesn't declare dependency installation; the scripts themselves include comments like 'pip install requests pandas'. This is not malicious but means dependencies must be installed manually. The smoke-test script expects an unlisted 'uv' CLI tool to exist (see environment note).
Credentials
Scripts optionally read SEMANTIC_SCHOLAR_API_KEY via environment; that is expected and proportionate for an API client. Registry metadata lists no required env vars; this is acceptable because the API key is optional in code, but users should be aware the key increases rate limits. No unrelated credentials or secrets are requested.
Persistence & Privilege
Skill does not request always:true, does not modify other skills or global agent config, and only writes its own output files. Autonomous invocation is allowed by default (platform normal) but combined with the rest of the footprint does not introduce unusual privilege.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install semantic-scholar - After installation, invoke the skill by name or use
/semantic-scholar - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of the Semantic Scholar skill.
- Enables searching, retrieving, and organizing scholarly metadata using Semantic Scholar APIs.
- Supports interactive search, batch fetching by IDs, recommendations from seed papers, and offline dataset workflows.
- Provides guidance on workflow selection: Graph API for lookups, Recommendations API for related work, Datasets API for offline needs.
- Includes best practices for efficient requests, pagination, authentication, and error handling.
- Offers example scripts for bulk paper harvesting, batch metadata fetch, recommendations, and smoke testing.
Metadata
Frequently Asked Questions
What is semantic-scholar?
Search, retrieve, and organize scholarly metadata with the Semantic Scholar APIs. Use when Codex needs to find papers or authors, build paper sets from compl... It is an AI Agent Skill for Claude Code / OpenClaw, with 347 downloads so far.
How do I install semantic-scholar?
Run "/install semantic-scholar" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is semantic-scholar free?
Yes, semantic-scholar is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does semantic-scholar support?
semantic-scholar is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created semantic-scholar?
It is built and maintained by Siyu Liu (@grenzlinie); the current version is v1.0.0.
More Skills