The Agent Skills Directory

[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as its primary function involves the agent ingesting and analyzing potentially untrusted user content such as business plans, board agendas, and project post-mortems.
Ingestion points: User-supplied data enters the agent's context through parameters like <plan>, <agenda>, <decision>, <assumption>, and <event> across multiple command definitions in SKILL.md and the sub-skill files in the skills/ directory.
Boundary markers: Absent. The instructions do not define delimiters (like triple backticks) or provide instructions for the agent to ignore or escape instructions that might be embedded within the provided user text.
Capability inventory: The agent is instructed to execute local Python scripts (scripts/decision_matrix_scorer.py and scripts/stakeholder_mapper.py), read local context files (company-context.md), and invoke other agent roles (e.g., [INVOKE:cfo]).
Sanitization: Absent. There is no evidence of input validation, escaping, or filtering of the user-provided content before it is processed by the agent or passed to tools.

executive-mentor