The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the 'graphifyy' Python package from PyPI using pip, pipx, or uv. This is the primary tool used for the skill's functionality.
[COMMAND_EXECUTION]: The skill executes several shell commands through the 'graphify' CLI, including building the graph ('graphify update'), querying structure ('graphify query'), and installing git hooks ('graphify hook install'). These operations are directed at the local repository and filesystem.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests and processes untrusted codebase data. Ingestion points: Untrusted source code files and user-provided arguments in $ARGUMENTS. Boundary markers: There are no explicit delimiters or warnings to ignore embedded instructions within the codebase or graph output. Capability inventory: The agent can execute shell commands via the 'graphify' tool and write data to the 'graphify-out/' directory. Sanitization: The skill does not implement explicit sanitization or validation of the codebase content before it is parsed by the graph tool or read by the agent.

save-tokens