The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external mission reports to influence the agent's next steps and user reporting.
Ingestion points: Data enters the context via get_report(mission_id) and user-provided prompts in plan_project and create_mission.
Boundary markers: The skill does not define specific delimiters or instructions to the agent to treat mission reports as untrusted data or to ignore embedded instructions within them.
Capability inventory: The skill facilitates powerful capabilities including spawning parallel agents (dispatch_mission) with access to the local filesystem and git repositories via isolated worktrees.
Sanitization: There is no evidence of sanitization, validation, or escaping of mission report content before it is processed by the orchestrating agent.

claude-devfleet