The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external OMERO servers, including image metadata, tags, comments, and tabular data, which could potentially contain malicious instructions if the server is compromised or contains untrusted content.
Ingestion points: Data retrieval methods documented in references/data_access.md, references/metadata.md, and references/tables.md (e.g., conn.getObject, ann.listAnnotations, table.read).
Boundary markers: None identified in the provided code examples or instructions.
Capability inventory: The skill documents local file writing (references/advanced.md, references/image_processing.md) and the ability to upload and execute scripts on the remote OMERO server (references/scripts.md).
Sanitization: No specific sanitization or validation of retrieved metadata strings is shown in the examples.
[REMOTE_CODE_EXECUTION]: The skill documents the OMERO.scripts framework in references/scripts.md, which allows users to upload Python scripts to the OMERO server for remote execution. While this is a core feature of the OMERO platform, it represents a significant capability for remote code execution.
[COMMAND_EXECUTION]: The documentation includes examples of interacting with the local file system using Python's os and open functions to save images and download data from the server, as seen in references/advanced.md and references/image_processing.md.

omero-integration