The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes common shell commands to manage its workflow. It uses git rev-parse --show-toplevel to identify the project root for storing metadata and openssl rand -hex 4 to generate unique article identifiers. Additionally, it uses mv to synchronize storage directory names with article filenames. These commands are consistent with the skill's purpose as a development-oriented documentation tool.
[PROMPT_INJECTION]: The skill implements a workflow that reads and scans project source code to gather implementation details for articles. This creates an attack surface for indirect prompt injection (Category 8), as malicious instructions or adversarial content embedded within the scanned repository files could potentially influence the agent's reasoning or output. This risk is inherent to the skill's function and is managed through structured data storage in 'materials.md' and user verification checkpoints.

article-write