The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the keeper-secrets-manager-cli package via pip. This is the official CLI tool provided by the vendor for secrets management.
[COMMAND_EXECUTION]: The skill utilizes the ksm exec command to execute subprocesses with secrets injected as environment variables. It also employs tmux sessions to maintain authentication state and handle interactive prompts during CLI operations.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection because it retrieves data from an external source (the Keeper Vault) and incorporates it into the agent's context.
Ingestion points: Data enters the context via ksm secret list, ksm secret get, and ksm interpolate as documented in SKILL.md and references/ksm-commands.md.
Boundary markers: The instructions do not define explicit boundary markers for the retrieved data, although they advise the agent to inspect metadata first and confirm actions with the user.
Capability inventory: The skill allows shell command execution (ksm exec), file writing (ksm interpolate), and network operations (ksm sync) as described across the reference files.
Sanitization: There is no explicit logic provided to sanitize or escape the content retrieved from the vault before processing.

keeper-secrets