The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests content from external files at ../satisfy-your-agent/SKILL.md and ../satisfy-your-agent/references/activities.md and treats it as instructions to follow.
Ingestion points: Reads content from files in a sibling directory using relative path traversal.
Boundary markers: No delimiters or warnings are present to distinguish instructions within the read files from the current agent context.
Capability inventory: The skill has the capability to read local files and modify the agent's behavioral state (specifically by stopping production work).
Sanitization: No validation or sanitization is performed on the content retrieved from the external files.
[PROMPT_INJECTION]: The instructions command the agent to 'stop' production work and 'not ask the user to choose another activity,' which functions as a behavioral override of the agent's standard task-oriented responses.

sya-reflect