The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git commands to audit repository history, retrieve commit SHAs, and manage submodule pointers as part of its evidence pass.
[DATA_EXFILTRATION]: The skill includes instructions to perform git push operations to synchronize management documents with the project's specification repository remote.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing untrusted data from multiple external repositories. * Ingestion points: The skill reads artifact files and repository state from sibling implementation repositories and the spec repository. * Boundary markers: There are no explicit markers or instructions to ignore embedded commands within the audited project files. * Capability inventory: The skill has the ability to write management files to the local file system and execute network-enabled git commands. * Sanitization: The skill does not describe specific filtering or escaping mechanisms for content extracted from project artifacts before interpolating them into generated documents.

hsdd-checkpoint