The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to manage and search for content from the source book.
Evidence: SKILL.md provides command templates using git clone, curl, grep, and sed for fetching and searching the repository's files.
[EXTERNAL_DOWNLOADS]: The skill fetches data from an external GitHub repository belonging to the author.
Evidence: Instructs the agent to download content from https://github.com/eternity4719/HowToLiveBetter to access the book's text and configuration files.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from a public repository, creating a potential surface for indirect prompt injection if the source content were compromised.
Ingestion points: The skill fetches multiple Markdown and HTML files from github.com/eternity4719/HowToLiveBetter and incorporates their contents into the agent's context.
Boundary markers: The instructions do not specify the use of delimiters or specific security headers when interpolating the fetched content into the response.
Capability inventory: The agent has access to git, curl, grep, and sed to interact with the downloaded data.
Sanitization: No sanitization or validation of the fetched content is described before it is processed by the AI.

life-decision-guide