The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted user input via the $ARGUMENTS variable to construct a briefing for a second agent. This creates an indirect prompt injection surface where instructions hidden in the user's task description could influence the behavior of the receiving agent.
Ingestion points: User-supplied input in the $ARGUMENTS variable referenced in SKILL.md.
Boundary markers: The handoff prompt uses Markdown headers such as ## Task and ## Context to structure the briefing, which provides logical separation but no explicit instruction to the receiving agent to ignore potentially malicious embedded content.
Capability inventory: The skill uses the paseo orchestration tool to spawn and initialize new agents with project context.
Sanitization: There is no evidence of sanitization or filtering of user-provided task descriptions before they are passed to the next agent.

paseo-handoff