The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted telemetry data which could be manipulated to influence the agent's reporting or proposal logic.
Ingestion points: Reads log entries from .work-state/coordinator/<project-slug>/profile-usage.jsonl (SKILL.md).
Boundary markers: Absent; there are no instructions provided to treat the telemetry data as untrusted or to use specific delimiters when processing it.
Capability inventory: The skill has the ability to write reports to profile-stats.md.
Sanitization: The skill does not specify any validation or sanitization of the telemetry fields (such as workflow or type) before including them in the rollup or using them to generate profile proposals.

coordinator-stats