The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill operates by reading and writing to an external issue tracker (a shared 'map'), which creates a surface for indirect prompt injection if the tracker content is modified by an external party.
Ingestion points: The agent fetches the 'map' issue body, child ticket bodies (research, prototype, grilling, task), and resolution comments from the repository's issue tracker (SKILL.md).
Boundary markers: There are no specific instructions or delimiters mentioned to isolate external ticket content or to instruct the agent to ignore potentially malicious instructions embedded within those tickets.
Capability inventory: The skill has the capability to create and close issues, wire dependencies between tasks, and spawn specialized sub-agents such as /research and /prototype to execute further actions based on ticket content (SKILL.md).
Sanitization: The skill does not define any sanitization, filtering, or validation steps for the data retrieved from the issue tracker before it is used to drive the agent's logic.

wayfinder