The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided task descriptions and passes them as context to specialized sub-agents and backlog management tools, creating a surface where embedded instructions could potentially influence downstream agent behavior.
Ingestion points: User-supplied task descriptions provided as arguments to the /orchestrate command in SKILL.md.
Boundary markers: The instructions lack explicit requirements for using delimiters or boundary markers when delegating tasks to sub-agents like python-cli-architect or spec-analyst.
Capability inventory: The skill calls the mcp__plugin_dh_backlog__backlog_add tool and delegates complex workflows to multiple specialized agents across different files.
Sanitization: No explicit sanitization or validation of the user's task description is performed before it is interpolated into delegation contexts or tool arguments.

orchestrate