The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by executing the git-why CLI binary to manage project context. This is the intended primary functionality of the skill, and commands are used to save, retrieve, and search local files.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to app.gitwhy.dev for cloud synchronization and GitHub Pull Request integration. These requests are directed to the vendor's own infrastructure to support shared visibility and team features.
[CREDENTIALS_UNSAFE]: The skill manages its own authentication credentials stored in the standard user directory path ~/.gitwhy/credentials. This is a routine implementation for CLI-based services and does not represent a credential leak or unsafe exposure.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it retrieves and renders user-authored engineering contexts (reasoning, stories, and decisions). While it reads external data into the agent's context, this is the core utility of the tool for knowledge retrieval.
Ingestion points: Data enters the context through gitwhy_get and gitwhy_search from local storage or the vendor's API.
Boundary markers: The instructions do not specify explicit delimiters to isolate retrieved context from current agent instructions.
Capability inventory: The skill can execute shell commands and perform network operations.
Sanitization: No specific sanitization or filtering of the retrieved markdown content is described.

gitwhy-context-saving