The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses the 'ghq' utility to clone external repositories into the local environment for analysis.
[COMMAND_EXECUTION]: The workflow involves standard shell commands such as 'date' for logging, 'mkdir' for file organization, and 'ghq' for source code acquisition.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it analyzes untrusted content from external repositories. Ingestion points: Files are downloaded from external sources using 'ghq get'. Boundary markers: There are no explicit instructions for the agents to disregard potentially malicious commands embedded in the source files. Capability inventory: The agent can create directories and write multiple markdown files to the workspace. Sanitization: No specific sanitization or validation logic is applied to the content extracted from the repositories.

learn