The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the gbrain CLI tool for querying, reading, and writing markdown files in the local media/articles/ directory.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes raw, untrusted article content without protective measures.
Ingestion points: Raw article text is loaded via gbrain get from the local file system.
Boundary markers: Absent. The skill does not specify any delimiters or safety prompts to isolate the ingested text.
Capability inventory: File system read and write operations provided by the gbrain toolset.
Sanitization: Absent. No filtering or validation of the article source text is performed before LLM processing.

article-enrichment