The Agent Skills Directory

[SAFE]: The skill operates as a set of instructional prompts for an AI agent. It does not contain executable code, remote dependencies, or obfuscated content. Findings from specific categories are detailed below:
[DATA_EXPOSURE]: The skill instructions include reading from local context files (e.g., .agents/em-context.md and .agents/reports/[name].md). This is a standard pattern for context-aware AI agents and does not constitute unauthorized data exposure as the agent is intended to process this localized information to provide tailored management advice.
[EXTERNAL_DOWNLOADS]: The references/sources.md file contains links to external articles on manager.dev and staysaasy.com. These are provided for the user's reference and are not automatically fetched or executed by the agent at runtime. The domains match the author's verified context.
[PROMPT_INJECTION]: The skill uses negative constraints (e.g., 'Do NOT use for standard underperformance') to guide the agent's appropriate application of the skill, rather than attempting to bypass safety filters or override system-level instructions.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests potentially untrusted data from local report files, it lacks dangerous capabilities (such as network access or shell execution) that could be exploited via indirect injection.

managing-high-performers