The Agent Skills Directory

[PROMPT_INJECTION]: The skill establishes a workflow for analyzing and validating existing note structures, creating an indirect prompt injection surface.
Ingestion points: Existing note data processed by schema_infer, schema_validate, and schema_diff (SKILL.md).
Boundary markers: The system utilizes structural YAML delimiters but provides no specific instructions for the agent to ignore potentially malicious commands within the note body during processing.
Capability inventory: File modification capabilities including write_note and edit_note (SKILL.md).
Sanitization: No methods for escaping or validating the content of processed notes are mentioned before the data influences schema generation.
[SAFE]: No hardcoded secrets, sensitive file exfiltration, or unauthorized network operations were identified.
[NO_CODE]: The skill is a documentation-only resource and does not include executable scripts, binaries, or external package dependencies.

memory-schema