The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions direct the user to install the graphify-ts package globally via NPM (npm i -g graphify-ts). This package is not associated with a verified organization or trusted vendor, posing a potential security risk if the package source is untrustworthy or compromised.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted source code to build a structural AST index, creating a surface for potential prompt injection. * Ingestion points: Processes all source files in the directory during the graphify build . command. * Boundary markers: None are present to distinguish between legitimate code structure and potentially malicious symbol metadata. * Capability inventory: The skill uses the Bash tool to execute the graphify CLI commands. * Sanitization: There is no mention of sanitization or validation of the extracted AST data before it is presented to the agent.

graphify