The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides links to documentation and code repositories hosted on well-known services.
Evidence: SKILL.md contains links to https://github.com/cursor/plugins/blob/main/pstack/ and https://cursor.com/docs/skills.
These links are for user guidance and target established vendor domains.
[COMMAND_EXECUTION]: The skill instructs both the user and the agent on the use of slash commands to drive agent workflows.
Evidence: Numerous slash commands like /setup-pstack, /poteto-mode, and /recall are listed in SKILL.md and references/recipes.md.
These commands are intended to invoke other agent skills rather than direct shell execution.
[INDIRECT_PROMPT_INJECTION]: The skill parses user questions to provide routing and help, which creates a potential attack surface.
Ingestion points: User chat messages are ingested and processed in SKILL.md to map them to specific help topics.
Boundary markers: The skill does not use delimiters (like XML tags) or specific prompts to the agent to treat user input as data rather than instructions.
Capability inventory: While this skill is informational, it recommends other skills that have the capacity to modify the file system, install plugins, and manage subagents.
Sanitization: No input validation or sanitization is performed on the user's queries before they are interpreted by the agent.

poteto-help