The Agent Skills Directory

[NO_CODE]: The skill consists entirely of Markdown and JSON files defining logic and test cases for an agent's conversational behavior. There are no scripts (Python, JavaScript, shell) or compiled binaries included.
[SAFE]: The content is focused on psychological and lifestyle advice derived from published literature. No evidence of prompt injection, data exfiltration, or malicious intent was found.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user prompts regarding personal issues (family pressure, anger), it lacks any high-risk capabilities such as file system writing, network requests, or command execution, rendering the surface area for this attack vector effectively non-existent.
Ingestion points: User prompts defined by triggers in SKILL.md.
Boundary markers: Explicitly defined in the B — Boundary section, limiting the scope to non-legal and non-debt scenarios.
Capability inventory: None. No scripts or tool calls are present.
Sanitization: Not applicable as no external system calls are made.

self-liberation