The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides explicit instructions to install a third-party dependency 'technical-writer' from an external GitHub repository (https://github.com/shubhamsaboo/awesome-llm-apps) using the npx skills add command. This introduces a supply chain risk as the source is not an official or trusted vendor.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands through the npx utility to manage, list, and install agent skills within the environment.
[DATA_EXPOSURE]: To ensure technical accuracy, the agent is instructed to read internal project configuration files such as *.tfvars, config.py, and pyproject.toml. This behavior grants the agent access to architectural metadata and environment-specific settings.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its core functionality of reading and processing untrusted data.
Ingestion points: Processes content from README.md, TODO.md, DESIGN.md, architecture-and-walkthrough.md, testing.md, and various code configuration files.
Boundary markers: Absent. The instructions do not employ delimiters or ignore-instructions wrappers when the agent reads external file content.
Capability inventory: The agent possesses the capability to modify the local filesystem (writing documentation) and execute shell commands via the skills toolset.
Sanitization: Absent. No logic is defined to validate or sanitize external content before it is incorporated into documentation or used to inform agent decisions.

project-documentation