The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes data from external project management resources (Agent Kanban tasks, repositories, and boards) which could contain malicious instructions embedded by external actors.
Ingestion points: Data is retrieved from the environment using realmroot toolbox get agent-kanban/tasks, realmroot toolbox get agent-kanban/repositories, and realmroot toolbox get agent-kanban/boards as described in SKILL.md.
Boundary markers: The instructions lack explicit boundary markers or directives for the agent to ignore potentially malicious instructions embedded within the retrieved task goals or architecture definitions.
Capability inventory: The skill possesses the capability to modify project state, create new resources, and assign agents to tasks using realmroot toolbox post and realmroot toolbox patch commands.
Sanitization: There are no instructions provided to validate, sanitize, or filter the content of the tasks before the agent incorporates them into the project architecture and dependency graph.

ak-plan