The Agent Skills Directory

[SAFE]: The skill operates entirely within the scope of its intended purpose: managing squad coordination and leader routing within the Multica platform. It uses the permitted 'multica' CLI tool for all operations and provides clear documentation and warnings regarding side effects.\n- [INDIRECT_PROMPT_INJECTION]: The skill manages a squad's coordination instructions, which are later injected into a squad leader's briefing. This creates a potential surface for indirect prompt injection if malicious data is provided as instructions. This is a core functional feature of the platform and is managed as intended.\n
Ingestion points: Squad instructions are ingested via the multica squad update command and recorded in the instructions field as described in SKILL.md.\n
Boundary markers: The platform provides a structured 'Squad Leader Briefing' that includes a specific 'Squad Instructions' segment to isolate this content.\n
Capability inventory: The skill and the resulting leader agent are restricted to the 'multica' CLI toolset.\n
Sanitization: No explicit sanitization of instruction text is performed within the skill; the platform handles the final prompt construction.

multica-squads