The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands through the terminal tool, including ls, find, and git commands for repository discovery and SHA tracking.
[EXTERNAL_DOWNLOADS]: The skill facilitates downloading entire codebases from user-provided URLs using git clone to a temporary directory.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external repositories, creating a surface for instructions embedded in source files to influence agent behavior.
Ingestion points: The agent reads manifest files (e.g., package.json, pyproject.toml), READMEs, and source code files using the read_file tool (SKILL.md).
Boundary markers: There are no explicit instructions or markers defined to separate the data read from external files from the agent's instructions, nor are there warnings to ignore embedded commands.
Capability inventory: The agent possesses terminal access (capable of executing shell commands) and write_file access across the file system (SKILL.md).
Sanitization: The instructions do not define any sanitization, filtering, or validation for the content retrieved from the analyzed repositories before it is processed by the agent.

code-wiki