The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill mandates the installation of 'OpenClawCLI' from https://clawhub.ai/. This domain is not an established trusted vendor or well-known service, which creates a risk of downloading and executing unverified or malicious binaries.
[COMMAND_EXECUTION]: The instructions guide the agent to perform shell operations, including pip install for multiple dependencies and the execution of a local Python script scripts/research.py with various flags.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external academic databases such as arXiv, PubMed, and Semantic Scholar, which could contain adversarial content.
Ingestion points: Data is retrieved from external APIs (titles, abstracts, and full-text PDFs) as described in SKILL.md.
Boundary markers: The skill lacks specified delimiters or instructions to ignore embedded commands within the retrieved research data.
Capability inventory: The skill has capabilities for file system writes (--output, --output-dir) and network access for querying APIs.
Sanitization: There is no mention of sanitizing or validating the retrieved academic content before processing.

academic-research-hub