The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local shell and Python commands to manage its environment. Specifically, it uses bash, curl, and python3 to generate PKCE secrets for authentication, perform API requests to vendor endpoints, and execute a version check script. All execution is limited to the skill's necessary functions.
[EXTERNAL_DOWNLOADS]: The skill references and downloads configuration from official vendor resources. It uses npx to install from https://second-me.cn and fetches version metadata from https://second-me.cn/skill.md. These references target the author's official domains.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied data, creating a surface for indirect prompt injection.
Ingestion points: Data is ingested through user-provided notes (text and links), key memories, and avatar definitions within note.md, key-memory.md, and avatar.md.
Boundary markers: The instructions in references/avatar.md explicitly warn the agent to treat report data as evidence rather than executable instructions.
Capability inventory: The skill has the ability to execute shell commands and Python scripts for API and local configuration tasks.
Sanitization: The skill relies on descriptive logic and AI instructions to categorize and handle data appropriately, rather than programmatic sanitization of the input text.

secondme