The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the mkit-cli binary from an external source using cargo install mkit-cli.
[COMMAND_EXECUTION]: The skill uses shell commands to drive the mkit CLI for repository management, key generation, and object inspection.
[DATA_EXFILTRATION]: The skill supports network protocols such as mkit+https, mkit+s3, and mkit+ssh and provides commands like mkit push that allow local repository content to be sent to external endpoints.
[CREDENTIALS_UNSAFE]: The skill manages private signing keys stored in .mkit/keys/ and interacts with the SSH_AUTH_SOCK environment variable for authentication, which are sensitive assets.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. 1. Ingestion points: Data enters the environment via mkit clone and mkit pull from remote repositories and mkit add for local files (SKILL.md). 2. Boundary markers: Absent; there are no instructions to the agent to ignore embedded commands or instructions within repository data. 3. Capability inventory: The skill can execute shell commands, write to the filesystem, and perform network operations via the mkit CLI (SKILL.md). 4. Sanitization: Absent; no validation or escaping of repository content is specified before processing.

mkit