The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary Python script, scripts/waypoint.py, executes git commands (git rev-parse --show-toplevel and git ls-files) via the subprocess module to resolve the repository root and identify tracked files. These calls use fixed argument lists and do not invoke a shell, protecting against command injection.
[INDIRECT_PROMPT_INJECTION]: The skill scans files in the repository to extract navigation markers, creating an attack surface where malicious files could attempt to influence the agent's behavior.
Ingestion points: The gather_code_blocks function in scripts/waypoint.py reads the content of all tracked files in the repository to identify Waypoint markers.
Boundary markers: No specific boundary delimiters are used to protect the agent from instructions embedded within the source files being analyzed.
Capability inventory: The skill possesses the ability to write to the filesystem through the block --write and manifest commands, which can modify existing source code files or create new documentation files.
Sanitization: The script parses data using regular expressions but does not perform sanitization on the role or description text extracted from code blocks before returning it to the agent.

waypoint