The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a Bash loop with awk and grep to process a configuration file located at ~/.superskills/knowledge.conf. This script searches for research material in directories specified by the user and is limited to local file system operations.
[PROMPT_INJECTION]: The skill involves processing content from external files, which creates an indirect prompt injection surface.
Ingestion points: Local files (.md, .txt, .html) discovered in directories listed in the user's knowledge base configuration.
Boundary markers: None identified; the skill does not use delimiters to isolate knowledge base content from instructions.
Capability inventory: Employs Bash, Grep, Read, Write, and Glob tools to interact with the environment.
Sanitization: There is no evidence of sanitization or validation of the content read from the files before it is used by the AI.

content-writer