The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill enables the agent to ingest untrusted data from an OMERO server, including image names, dataset descriptions, and annotation values, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: The skill provides methods to retrieve various forms of metadata and structured data in references/data_access.md, references/metadata.md, and references/tables.md (e.g., conn.getObjects, image.listAnnotations, table.read).
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the provided templates.
Capability inventory: The skill possesses significant capabilities, including file system writes (references/advanced.md, references/metadata.md), server-side object creation (references/image_processing.md), and object deletion (references/advanced.md).
Sanitization: No sanitization or validation of the retrieved text content is implemented before processing.
[SAFE]: The skill identifies and utilizes the legitimate omero-py Python package for interacting with the OMERO platform.
[SAFE]: Credential management instructions in references/connection.md promote the use of environment variables and configuration files, avoiding hardcoded secrets in the code logic.
[SAFE]: File management operations, such as downloading and uploading microscopy data or analysis results, are consistent with the skill's primary purpose and are implemented using standard Python library functions.

omero-integration