The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process multiple user-controlled files within a 'learning workspace' to determine its next actions and instructional logic. This creates a surface where embedded instructions in these files could influence agent behavior.
Ingestion points: The agent is instructed to read MISSION.md, STUDY-PLAN.md, RESOURCES.md, GLOSSARY.md, NOTES.md, and content from the lessons/, exercises/, and learning-records/ directories (referenced in SKILL.md).
Boundary markers: The instructions do not define explicit boundary markers or delimiters (e.g., XML tags or specific 'ignore instructions' warnings) when interpolating the content of these files into its context.
Capability inventory: The skill has the capability to write and update markdown files in the local filesystem to maintain learning state.
Sanitization: There are no instructions for sanitizing or validating the content of the ingested files before processing them for instructional logic.

learning-coach