Start a NEW task in the local Kimi Code agent — a separate CLI coding agent with its own file/shell tools and its own context window. Blocks until kimi settles, then returns only its final result plus stats, prefixed [session: ]; continue that session later with kimi_reply. Good for: a second opinion from a different model, work kept out of this context, or parallel investigation. Caution: with approval_mode 'yolo' (the server default, ACP) kimi edits files and runs shell commands as your user inside cwd without asking. Print mode also executes tools without confirmation (and forbids combining -p with -y).

Send a new turn to an existing kimi session that is not executing right now — including one that timed out or was cancelled: the session survives in kimi's own store, so resume it here instead of restarting with kimi. kimi still has its prior turns (but never this conversation), so the follow-up can be short. Survives restarts of this server. For a turn still running under 'acp', use kimi_send instead.

List the models this kimi installation has configured right now — read live from kimi provider list --json. Use it to pick a model value for kimi / kimi_reply. Starts no task. This is the configured catalog, not a paid probe of every endpoint.

Deliver a message into a kimi turn that is executing right now. The default transport 'acp' keeps the turn reachable; 'print' runs cannot be reached, and a session that already finished takes kimi_reply, not kimi_send. Returns immediately; kimi's reaction appears in the answer of the kimi/kimi_reply call still waiting on that turn. kimi_running lists reachable sessions. Note: the adapter reports the first result envelope; a follow_up queued behind the current turn is delivered but its answer is not awaited by the original call.

List kimi turns executing at this moment — the ones kimi_send can reach — with session id, working directory, elapsed time, and messages already sent in. Only acp-transport runs appear; 'print' runs are unreachable mid-run. For past sessions use kimi_sessions.

List all kimi sessions started through this server, newest first, with their working directory — running or finished, including runs that timed out. Use it to recover an id for kimi_reply. For turns still executing (kimi_send targets), use kimi_running.

Read a bounded snapshot of a kimi session's visible conversation from kimi's native agents/main/wire.jsonl. Does not send anything into the run, does not take the session lock, and does not wait for kimi to finish. Returns JSON: session, state (active if this process is running it, otherwise unknown — sessions stay resumable so 'completed' is not claimed), items (user / assistant / tool / gap), an opaque cursor to resume even at the current EOF, has_more, truncated_tail. Thought/reasoning/encrypted content are omitted. User text comes only from context.append_message with role=user — if that stream has none, the page says so rather than inventing prompts from session metadata. Caps: limit (default 50, max 200), max_chars per item (default 4000, max 16000), and 100000 chars on the whole JSON page.