Zen's CLI surface is intentionally small: the TUI plus 23 subcommands. Heavy capabilities (search, research, dispatch, briefs) run autonomously through the background scheduler and distill loop rather than as manual commands — drop notes into ~/.zen/vault/inbox/ and the system works on them.
Command Description
zenLaunch TUI (interactive agent session; first run initializes ~/.zen/) zen versionShow version information zen --helpShow full help with all commands
Command Description
zen chatInteractive LLM chat (non-TUI) zen session startStart an agentic session zen session statusShow session status zen session listList sessions zen session archiveArchive a session
Command Description
zen wiki listList wiki pages zen wiki show <page>Show a wiki page zen wiki reindexRebuild the knowledge index (FTS5 + embeddings; also heals legacy non-NFC alias rows; --dry-run preview, --fts-only resync) zen wiki lintLint wiki (orphan pages, broken links, stale claims) zen wiki distillRun the consolidation pipeline (inbox → wiki) zen wiki rebuild-memoryRebuild the gateway memory store from Markdown + session archives (requires running daemon) zen wiki loop runRun one self-learning consolidation cycle now zen wiki loop statusLoop status (--json includes latency metrics) zen wiki loop gapsShow pending gaps and user questions zen wiki loop enable / disableToggle the background loop
Command Description
zen agent listList registered agents zen skill list --jsonList learned skills zen skill run <name>Run a skill zen skill progressShow skill learning progress zen skill show <name>Show a skill's details zen skill precipitateStage a distilled skill draft zen skill confirm <name>Confirm a staged skill zen discover runRun one self-learning cycle (stages 5b/5c/5d) zen discover stage / queueStage and inspect improvement hypotheses zen discover confirm / rejectReview staged hypotheses zen discover reportLearning metrics (routing, stalls, calibration, replay) zen discover arenaRun distill regression gate vs baselines zen discover calibrate --writeDerive decision thresholds from labeled data zen discover label status / next / setLabel supply workflow
Command Description
zen serve startStart gateway daemon (--foreground, --http for loopback HTTP carrier, --mcp for stdio MCP mode) zen serve statusDaemon health zen serve stopStop daemon zen serve testTest MCP connectivity zen serve install / uninstallManage macOS launchd auto-start zen doctorSystem health: 10 liveness probes (--json for scripts) zen clean cacheClean caches (ungated) zen clean all / trashDestructive cleanup — asks for confirmation unless --yes zen logs <service>View structured logs zen auditAudit log operations zen sandbox testVerify sandbox isolation works
Command Description
zen workspace initInitialize ~/.zen/ workspace structure zen config showShow effective configuration (layers merged) zen provider listList available LLM providers zen provider test <name>Test a provider connection zen auth listList stored credentials zen modelModel metadata + routing info
Command Description
zen starter <template>Generate project scaffold from template zen wps <action>Work process utilities zen habit log / listHabit tracking zen goal create / list / statusGoal management zen plugin list / install / enable / disable / rehashPlugin management (WASM tools, MCP)
# Daily workflow — all through the TUI
zen
# Or scripted
zen doctor # health check
zen wiki loop status # is the learning loop alive?
zen wiki distill # drain the inbox now
zen discover report # what did the system learn?
# Maintenance
zen clean cache
zen serve status
Full reference: zen --help for the most up-to-date list of commands and flags.