The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill's documentation includes instructions to install the parallel-cli utility via curl -fsSL https://parallel.ai/install.sh | bash. This is a standard installation method for this tool, which is used for broad academic literature discovery.
[COMMAND_EXECUTION]: The scripts/generate_pdf.py file uses subprocess.run() to execute the pandoc system command. This execution is necessary for the skill's intended purpose of converting markdown documents into formatted PDF files.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to established academic infrastructure services such as PubMed, arXiv, bioRxiv, and CrossRef to retrieve research findings and verify citation metadata.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted data from academic search results. 1. Ingestion points: External data enters the context through parallel-cli search and various scholarly APIs. 2. Boundary markers: No specific delimiters or "ignore instructions" warnings are included in the synthesis workflow. 3. Capability inventory: The skill possesses capabilities for subprocess execution (scripts/generate_pdf.py), file writing, and network operations (scripts/verify_citations.py). 4. Sanitization: No content validation or escaping is applied to the retrieved paper content before it is processed by the agent.

literature-review