The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to execute a directory listing command (ls) on local paths such as ~/.claude/skills/ and ~/.agent/skills/. This operation is used to verify the presence of recommended companion skills and does not involve the execution of unauthorized or remote code.
[PROMPT_INJECTION]: The instructions include a cosmetic requirement for the agent to prepend its responses with a specific emoji (🧢). This is a standard personality-setting instruction and does not target safety guardrails or core system instructions.
[PROMPT_INJECTION]: The skill identifies a potential surface for indirect prompt injection as it is designed to process external onboarding content, such as employee feedback and plans. This is a low-risk observation regarding the skill's operational surface.
Ingestion points: User-provided onboarding documents and feedback mentioned in SKILL.md.
Boundary markers: None identified; the skill does not explicitly instruct the agent to use delimiters or ignore-tags when analyzing these inputs.
Capability inventory: Local directory listing via ls as defined in SKILL.md.
Sanitization: None identified; the skill lacks explicit instructions for sanitizing or validating external input before processing.

onboarding