The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes the @tnnevol/halo-cli tool via npx to perform operations on the Halo CMS. User-provided input for article titles and Markdown content is passed as command-line arguments to the CLI tool.
[EXTERNAL_DOWNLOADS]: The skill downloads the @tnnevol/halo-cli package from the npm registry. The setup documentation indicates that the tool may download platform-specific binaries during execution to handle API requests. These resources are associated with the skill author.
[PROMPT_INJECTION]: The skill retrieves and displays post data from an external Halo CMS instance, creating a surface for indirect prompt injection. Malicious instructions contained within a blog post could potentially influence the agent when performing retrieval actions.
Ingestion points: Content retrieved from the Halo RESTful API and displayed in the agent context via the get and list actions.
Boundary markers: No explicit delimiters or instructions are used to separate external content from the agent's internal logic.
Capability inventory: The skill possesses the capability to execute shell commands using npx for all defined actions.
Sanitization: The skill's security guidelines mention that sensitive values within API responses are automatically sanitized.

halo