AgentKit
ReferenceCLI commands
BetaYou are reading docs for the beta channel (2.4.1-beta.1) — updated automatically on every beta release.Switch to stable →

CLI commands

ak skill

Manage per-skill runtime envs

(For catalog browsing - list, install, remove - see ak skills. This command operates on a single per-skill runtime env.) Commands for managing the per-skill Python/Node runtime envs that ak run activates before executing a skill. Each subcommand operates on a single <kit>/<skill> reference.

Who it's for: Power devs scripting env hygiene; non-technical users via the matching dashboard surface.

When to use it: Run ak skill verify after a fresh checkout to confirm all envs are healthy; ak skill install after editing skill.yaml to populate the env; ak skill repair when verify reports corrupt; ak skill upgrade after changing the runtime block.

What changes on disk: Read-only at this group level; subcommands document their own effects under ~/.agentkit/skills/<kit>/<skill>/ and ~/.agentkit/cache/<hash>/.

Examples

ak skill verify marketing/youtube       # report health (ok|missing|corrupt|unknown)
ak skill install marketing/youtube      # create or refresh the env
ak skill repair marketing/youtube       # rebuild a corrupted env
ak skill upgrade marketing/youtube      # re-resolve after editing skill.yaml
ak skill remove marketing/youtube       # tear down the per-skill record

Flags

FlagDescription
-h, --helphelp for skill
--jsonEmit machine-readable JSON (implies --no-interactive)
--no-interactiveDisable interactive prompts (CI-safe)
-q, --quietSuppress non-error output on stderr
-V, --verboseExtra diagnostic output on stderr (loses to --quiet)
-y, --yesAssume yes for all prompts

Output modes

ModeBehavior
prettydefault on TTY (colors, ASCII markers)
plainauto when stdout piped or --no-interactive
json--json (single-object envelope, NDJSON-safe)

Exit codes

CodeMeaning
0success
1runtime error
2invalid flags
3user-cancel (SIGINT, prompt-cancel)

On this page