The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs users to install a Python wrapper from an unverified third-party GitHub repository (https://github.com/mcmero/labarchives-py). This library is a core dependency but originates from a source outside of the trusted organization list.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the ingestion of external data retrieved via the LabArchives API.
Ingestion points: Scientific entries, comments, and notebook metadata are fetched using methods in scripts/notebook_operations.py and scripts/entry_operations.py.
Boundary markers: The scripts do not use explicit delimiters or "ignore" instructions when processing fetched HTML or XML content, which could allow malicious instructions embedded in a notebook to influence the agent.
Capability inventory: The skill includes file-writing capabilities for local backups and network communication via the requests library.
Sanitization: There is no evidence of data sanitization or structural validation for the content retrieved from the API before it is displayed or written to disk.
[DATA_EXFILTRATION]: The skill enables automated, large-scale retrieval and local storage of research notebooks and attachments via scripts/notebook_operations.py. While intended for legitimate backups, this functionality provides a capability for data exfiltration that should be monitored for unauthorized exports.

labarchive-integration