The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches a compiled binary from the author's GitHub repository (github.com/mnhkahn/cyeam-cli) during the installation phase.
[REMOTE_CODE_EXECUTION]: The installation process involves downloading a remote binary and executing it, which constitutes a remote code execution risk if the source or transport is compromised.
[COMMAND_EXECUTION]: The skill uses shell commands to install and run the cyeam CLI tool. The installation instructions explicitly use the sudo command to move the downloaded binary into a system directory (/usr/local/bin/), which requires administrative privileges.
[PROMPT_INJECTION]: There is a risk of indirect prompt injection because the skill fetches HTML content from external OneDrive notes and renders it as Markdown. If these notes contain adversarial instructions, they could influence the agent's behavior.
Ingestion points: OneDrive HTML notes via cyeam cnote get command (SKILL.md).
Boundary markers: None identified in the prompt templates to isolate external content.
Capability inventory: The skill can execute shell commands and interact with the OneDrive API.
Sanitization: No explicit sanitization or filtering of the fetched content is mentioned before it is processed by the agent.

cnote