Skip to content

commands

github-actions[bot] edited this page Nov 16, 2026 · 1 revision

Commands

Look up the exact command, arguments and options for reading meetings, archiving recordings or managing the Zoom schedule. Stored meeting IDs differ from remote Zoom IDs; archive destinations are explicit local directories. Start with installation or a task guide when connecting for the first time.

Global options

These options apply to every command.

Option What it does
-V, --version output the version number.
-v, --verbose more detail in what is shown: -v ids, -vv everything we know. Default: 0.
--json machine-readable output: one JSON value on stdout; --json=id,text keeps only those fields.
--jsonl machine-readable output: one JSON object per line, for streaming and jq.
--quiet diagnostics off; a failure is still said.
--trace the connection's own log lines on stderr — never content.
--debug on a failure, its whole record — cause and stack — in place of the short line.
--timeout <duration> give up on the whole command after this — 30s, 2m, 500ms.
--offline answer from what was recorded and never connect; fails if nothing was.
--no-input never prompt or open interactive login; piped input remains available.
--max-input-bytes <bytes> maximum buffered input bytes (default: 16777216).
--max-output-bytes <bytes> maximum machine output bytes (default: 4194304; 0 disables).
--brief found items as reference, title, time, author and excerpt only.
--full found items with their whole text and every provider field.
--compact one line per item for an agent: reference first, text last, paging and next steps as # lines.
--dry-run preview parsed arguments and permissions before running the action.
--yes go ahead without the question an ask level puts before a write.
--record keep this run even where the config says record false; record_input false leaves the typed words out.
--no-record do not keep this run, even if it fails.

zm runs

recorded runs — what this tool did, and when

zm runs list

recorded runs, newest first

zm runs list [query] [options]
Argument What it is
query optional field:value terms that narrow the list — the fields are below.
Option What it does
--limit <n> how many to show.
--page <n> which page, starting at 1.

Returns: run — run_id, command, started_at, status, cli_version, profile, completed_at, requests, and 23 more; every field: zm commands info runs list

Example: zm runs list --limit 10

zm runs show

one run: what it was, how it ended, and every event it logged

zm runs show <run-id> [options]
Argument What it is
run-id required an id from zm runs list.
Option What it does
--problems only failed responses, warnings and crashes.

Returns: run details — run_id, command, started_at, status, cli_version, directory, event_count, events, and 22 more; every field: zm commands info runs show

Example: zm runs show run-101

zm runs path

the directory holding one run

zm runs path <run-id>
Argument What it is
run-id required an id from zm runs list.

Returns: run location — path; every field: zm commands info runs path

Example: zm runs path run-101

zm doctor

the state this installation is in, without connecting unless --online

zm doctor [options]
Option What it does
--online also connect once and read the account; sends nothing.

Example: zm doctor, zm doctor --json

zm doctor reports

problem reports

zm doctor reports show

what a problem report holds; writes nothing

zm doctor reports show

Example: zm doctor reports show

zm doctor reports create

write a problem report to a file, and say where to send it

Changes local data.

zm doctor reports create [options]
Option What it does
--run <id> the run the report is about; the newest failed one if not given.
--output <file> where to write it; a new file in this directory if not given.

Example: zm doctor reports create --output ./problem-report.json

zm mcp

serve this profile to an agent over MCP and set it up — claude mcp add zm -- zm <profile> mcp serve

zm mcp serve

serve this profile to an agent over MCP, on stdin and stdout — claude mcp add zm -- zm <profile> mcp serve; zm mcp serve --profiles a,b or --profiles all serves those profiles from one server and each tool asks which

Changes remote Zoom data.

zm mcp serve [options]
Option What it does
--permission <key=level> make a permission stricter for this server only, like meetings.cancel=deny; it never loosens the profile; repeat for more keys.
--profiles <names|all> serve several profiles from one server instead of one: names, comma-separated, or all for every profile.
--http serve over HTTP on 127.0.0.1 for ChatGPT and Claude in the browser, behind your tunnel; the profile's permissions decide.
--port <port> the local port for --http (default 8765).
--public-url <url> the tunnel's https address the browser apps use, e.g. https://.ts.net or https://.ts.net/tg.

Example: zm personal mcp serve, zm personal mcp serve --profiles work,personal, zm personal mcp serve --http --public-url https://laptop.example.ts.net

zm mcp revoke

forget every login mcp serve --http gave a browser app, of this profile or of the server --profiles names; each app must log in again

Changes local data.

zm mcp revoke [options]
Option What it does
--profiles <names|all> serve several profiles from one server instead of one: names, comma-separated, or all for every profile.

Example: zm personal mcp revoke, zm personal mcp revoke --profiles all

zm mcp server

start, stop and check mcp serve --http in the background; status shows the login code for a browser app

zm mcp server start

start mcp serve --http apart from this shell and answer once it listens, with its login code

Changes local data.

zm mcp server start [options]
Option What it does
--port <port> the local port (default 8765).
--public-url <url> the tunnel's https address the browser apps use, e.g. https://.ts.net or https://.ts.net/tg.
--permission <key=level> make a permission stricter for this server only; repeat for more keys.
--profiles <names|all> the server of several profiles that mcp serve --profiles serves: names, comma-separated, or all.

Example: zm personal mcp server start --public-url https://laptop.example.ts.net

zm mcp server stop

stop mcp serve --http — only one that mcp server start started

Changes local data.

zm mcp server stop [options]
Option What it does
--profiles <names|all> the server of several profiles that mcp serve --profiles serves: names, comma-separated, or all.

Example: zm personal mcp server stop

zm mcp server status

whether mcp serve --http runs, on which port and address, the login code for a new browser app, and the apps logged in

zm mcp server status [options]
Option What it does
--profiles <names|all> the server of several profiles that mcp serve --profiles serves: names, comma-separated, or all.

Example: zm personal mcp server status --json

zm mcp config

print the mcpServers entry for Claude Desktop, Cursor and others, with full paths; writes nothing

zm mcp config [options]
Option What it does
--permission <key=level> make a permission stricter for this server only, like meetings.cancel=deny; it never loosens the profile; repeat for more keys.
--profiles <names|all> serve several profiles from one server instead of one: names, comma-separated, or all for every profile.

Example: zm personal mcp config, zm personal mcp config --profiles all

zm mcp add

add this profile's local MCP server to Codex or Claude Code

Changes local data.

zm mcp add <client> [options]
Argument What it is
client required codex or claude-code.
Option What it does
--allow-writes acknowledge that this profile offers writing tools.
--permission <key=level> make a permission stricter for this server only, like meetings.cancel=deny; it never loosens the profile; repeat for more keys.

Example: zm personal mcp add codex --allow-writes

zm mcp check

check this profile's local MCP handshake and tool list

zm mcp check [options]
Option What it does
--permission <key=level> make a permission stricter for this server only, like meetings.cancel=deny; it never loosens the profile; repeat for more keys.

Example: zm personal mcp check

zm config

local Zoom configuration

zm config show

show local configuration

zm config show

Example: zm config show --json

zm config set

set a configuration value

Changes local data.

zm config set <key> <value>
Argument What it is
key required setting name, such as zoom.client_id.
value required new value.

Example: zm config set files.timezone Europe/Madrid --json

zm config unset

remove a setting, or a part of the file doctor reports a problem in

Changes local data.

zm config unset <key>
Argument What it is
key required setting name, such as files.folder, or the key a doctor fix names.

Example: zm config unset files.folder

zm config migrate

write each setting the file lacks at its default, keeping yours

Changes local data.

zm config migrate

Example: zm config migrate

zm app

what to enter when you make the Zoom app zm signs in with

zm app

Returns: zoom app setup — marketplace_url, guide_url, app_type, management, oauth, redirect_url, allow_list, scopes, and 1 more; every field: zm commands info app

Example: zm app, zm app --json

zm meetings

stored meetings and their records

zm meetings list

meetings, newest first: the service's and the stored ones, kept in the store

zm meetings list [query] [options]
Argument What it is
query optional field:value terms that narrow the list — the fields are below.
Option What it does
--backend <archive|server|both> where to read: the store (archive), the service (server), or both, kept in the store (default: both).
--limit <n> meetings per page; the profile's defaults.limit when absent.
--page <n> which page, starting at 1.

Returns: meeting — ref, id, provider_id, provider, account, title, url, started_at, and 12 more; every field: zm commands info meetings list

Next usually: zm meetings show <meeting>, zm meetings transcript <meeting>, zm meetings summary <meeting>

Example: zm personal meetings list --json

zm meetings show

show a meeting, read again from the service with its people, transcript and documents

zm meetings show <meeting> [options]
Argument What it is
meeting required the stored meeting's number, its meeting: ref, or the service's id (provider_id).
Option What it does
--backend <archive|server|both> where to read: the store (archive), the service (server), or both, kept in the store (default: both).

Returns: meeting details — ref, id, provider_id, provider, account, title, url, started_at, and 21 more; every field: zm commands info meetings show

Next usually: zm meetings transcript <meeting>, zm meetings summary <meeting>, zm meetings invitation <meeting>

Example: zm personal meetings show 1 --json

zm meetings transcript

show current meeting transcripts

zm meetings transcript <meeting> [options]
Argument What it is
meeting required stored meeting id.
Option What it does
--history include superseded transcript versions.

Returns: meeting transcript — ref, meeting, source, language, superseded_at, created_at, lines, untrusted; every field: zm commands info meetings transcript

Example: zm personal meetings transcript 1 --json

zm meetings summary

show stored meeting summaries

zm meetings summary <meeting>
Argument What it is
meeting required stored meeting id.

Returns: meeting summary — id, meeting, source, title, overview, sections, next_steps, content, and 4 more; every field: zm commands info meetings summary

Next usually: zm meetings show <meeting>, zm meetings transcript <meeting>, zm meetings invitation <meeting>

Example: zm personal meetings summary 1 --json

zm meetings export

export a meeting as Markdown or one transcript as WebVTT

Changes local data.

zm meetings export <meeting> [options]
Argument What it is
meeting required stored meeting id.
Option What it does
--format <format> export format. One of: markdown, vtt. Default: markdown.
--transcript-id <id> export a specific transcript version.
--output <file> write content to a new file through the host's writer.

Returns: meeting export — meeting_id, format, filename, content, untrusted, output; every field: zm commands info meetings export

Example: zm personal meetings export 1 --format markdown --json

zm meetings people

find meeting participants by name or email

zm meetings people <query> [options]
Argument What it is
query required a name or an email address.
Option What it does
--limit <n> most matching observations per page, 1–1000 (default 50).
--scan-limit <n> most participant observations to inspect, 1–10000 (default 1000).
--max-read-bytes <bytes> most stored name and email bytes to read (default 1048576).
--cursor <cursor> continue with the same name query and selected account.

Returns: meeting participants — schema_version, kind, account_id, query, items, ambiguity, coverage, page; every field: zm commands info meetings people

Example: zm personal meetings people "Alice Example" --json

zm meetings sync

bring the stored meetings up to date from the connected account, oldest first; a run that stops before its deadline keeps what it saved and the next one goes on from there

Changes local data.

zm meetings sync [options]
Option What it does
--since-time <time> fetch meetings starting at or after this ISO 8601 date or time, or 30m, 2h, 7d ago.
--lookback-days <days> revisit recent occurrences for late transcripts (0–31).
--rescan read the last 30 days again, not only what came after the last sync.

Returns: meeting ingestion — meetings, participants, transcript_rows, summaries, warnings, timezone; every field: zm commands info meetings sync

Example: zm personal meetings sync --dry-run --json

zm meetings import

import downloaded meeting transcripts

Changes local data.

zm meetings import <folder>
Argument What it is
folder required folder of transcripts downloaded from the provider.

Returns: meeting ingestion — meetings, participants, transcript_rows, summaries, warnings, timezone; every field: zm commands info meetings import

Example: zm personal meetings import ./downloaded-transcripts --json

zm meetings repair

preview filling in what older builds left out of stored transcripts (their language); --yes stores it

Changes local data.

zm meetings repair [options]
Option What it does
--yes store the detected languages; otherwise only preview.

Returns: meeting repair — account_id, applied, transcripts, languages; every field: zm commands info meetings repair

Example: zm personal meetings repair --json

zm meetings create

create a meeting on the service

Changes remote Zoom data.

zm meetings create [options]
Option What it does
--title <title> the meeting's title.
--kind <kind> an instant or a scheduled meeting. One of: instant, scheduled.
--start-time <time> when it starts: an ISO time with its offset, 2026-11-20T10:00:00+01:00.
--duration-minutes <minutes> how long it lasts, 1–1440.
--timezone <zone> its IANA time zone, Europe/Madrid.
--agenda <text> its agenda.
--setting <name=value> a setting of this service; repeat it for more: this service has none.

Returns: meeting change — action, provider, account, applied, id, occurrence_id, change, meeting, and 1 more; every field: zm commands info meetings create

Example: zm personal meetings create --json

zm meetings update

change a meeting on the service

Changes remote Zoom data.

zm meetings update <meeting> [options]
Argument What it is
meeting required the service's id of the meeting.
Option What it does
--title <title> the meeting's title.
--kind <kind> an instant or a scheduled meeting. One of: instant, scheduled.
--start-time <time> when it starts: an ISO time with its offset, 2026-11-20T10:00:00+01:00.
--duration-minutes <minutes> how long it lasts, 1–1440.
--timezone <zone> its IANA time zone, Europe/Madrid.
--agenda <text> its agenda.
--setting <name=value> a setting of this service; repeat it for more: this service has none.

Returns: meeting change — action, provider, account, applied, id, occurrence_id, change, meeting, and 1 more; every field: zm commands info meetings update

Example: zm personal meetings update 123456789 --title "Example review" --dry-run --json

zm meetings cancel

cancel a meeting on the service

Changes remote Zoom data.

zm meetings cancel <meeting> [options]
Argument What it is
meeting required the service's id of the meeting.
Option What it does
--occurrence-id <id> cancel only this occurrence of a repeating meeting.

Returns: meeting change — action, provider, account, applied, id, occurrence_id, change, meeting, and 1 more; every field: zm commands info meetings cancel

Example: zm personal meetings cancel 123456789 --dry-run --json

zm meetings invitation

the invitation text the service writes for a meeting

zm meetings invitation <meeting>
Argument What it is
meeting required the service's id of the meeting.

Returns: meeting invitation — id, invitation, untrusted; every field: zm commands info meetings invitation

Example: zm personal meetings invitation 123456789 --json

zm meetings scheduled

discover candidate meeting series from one bounded provider page

zm meetings scheduled [options]
Option What it does
--since-time <time> window start, an ISO date or time, local without a zone (default now).
--before-time <time> exclusive window end, an ISO date or time; a date alone covers that day (default seven days after start).
--limit <n> series candidates in one provider page, 1–100 (default 100).
--max-read-bytes <bytes> normalized provider-page bytes, at most 1048576 (default 65536).
--bytes <bytes> answer item UTF-8 JSON bytes, 2048–1048576 (default 65536).
--cursor <cursor> continue with the same absolute window and returned page cursor.

Returns: meeting series inventory — schema_version, kind, scope, after, until, items, page, fingerprint, and 2 more; every field: zm commands info meetings scheduled

Example: zm personal meetings scheduled --dry-run --json

zm meetings projects

project-name evidence in stored meetings

zm meetings projects evidence

find cited project-name and alias evidence in the local meeting index; does not refresh it or confirm membership

zm meetings projects evidence <project> [query] [options]
Argument What it is
project required existing project name, key or explicit alias.
query optional exact ID fields and a UTC started_at range, joined by AND.
Option What it does
--limit <n> hits in one name page, 1 through 1000 (default 100).
--cursor <cursor> continue with the same project and frozen selection from next.
--max-read-bytes <bytes> word-read byte budget, at most 67108864 (default 4194304).

Returns: meeting project evidence — schema_version, kind, account_id, query, project, term, hits, association, and 5 more; every field: zm commands info meetings projects evidence

Example: zm personal meetings projects evidence RES "identity_id:12" --json

zm meetings projects scan

suggest existing projects from one bounded page of stored meeting evidence

zm meetings projects scan <ref> [options]
Argument What it is
ref required canonical parent meeting reference in the selected account.
Option What it does
--limit <n> existing project candidates, 1–10 (default 10).
--cues <n> transcript cues, 1–100 (default 20).
--bytes <bytes> answer item UTF-8 JSON bytes, 2048–65536 (default 16384).
--max-read-bytes <bytes> stored text bytes per read, 1–65536 (default 65536).
--cursor <cursor> continue from evidence.coverage.next_cursor with the same meeting.

Returns: meeting project candidates — schema_version, kind, source, input_fingerprint, vocabulary_fingerprint, association, items, limits, and 2 more; every field: zm commands info meetings projects scan

Example: zm personal meetings projects scan meeting:1/2 --json

zm meetings prepare

prepare cited context for one upcoming series

zm meetings prepare <series-id> [options]
Argument What it is
series-id required exact provider series ID.
Option What it does
--context-source <id=cursor> authorized communications selection ID, optionally ID=cursor; repeat. Default: ``.
--context-query <words> literal words to find within the selected communications.
--context-since-time <time> communications window start, an ISO date or time or 7d ago (default fourteen days before end).
--context-before-time <time> exclusive communications end, an ISO date or time (default preparation start).
--project <project> include unconfirmed word evidence for one existing project name, key or alias.
--since-time <time> window start, an ISO date or time, local without a zone (default now).
--before-time <time> exclusive window end, an ISO date or time; a date alone covers that day (default seven days after start).
--history-since-time <time> history start, an ISO date or time or 30d ago (default thirty days before start).
--meetings <n> prior occurrences and separate related-meeting limit, 1–20 (default 5).
--cues <n> cues per occurrence, 1–100 (default 10).
--context-limit <n> entries per context category, 1–100 (default 20).
--max-read-bytes <bytes> stored/normalized bytes per read, at most 4194304 (default 262144).
--bytes <bytes> preparation item JSON bytes, 4096–4194304 (default 65536).

Returns: meeting preparation — schema_version, kind, account_id, upcoming, series, history_window, meetings, related, and 6 more; every field: zm commands info meetings prepare

Example: zm personal meetings prepare 123 --dry-run --json

zm meetings link

link a meeting to an event, or a participant to a person; --yes applies

Changes local data.

zm meetings link <reference> [options]
Argument What it is
reference required meeting reference, such as meeting:1/1.
Option What it does
--event-id <id> stored event id, or none to unlink.
--expected-event-id <id> with --event-id: compare current event id, or none.
--participant-id <id> with --person: participant id from this meeting.
--person <uid> with --participant-id: explicit stored person UID.
--source <accounts> explicit additional account ids authorized for target-person lookup.
--expected-person <uid> compare current person UID, or none.
--yes apply this local change once.
--dry-run preview even when --yes is present.

Returns: meeting association — scope, account_id, meeting_id, change, applied, result; every field: zm commands info meetings link

Example: zm personal meetings link meeting:1/1 --event-id 1 --json, zm personal meetings link meeting:1/1 --participant-id 1 --person 1 --json

zm meetings transcribe

preview transcribing a recording on this machine; --yes adds the transcript

Changes local data.

zm meetings transcribe <meeting> <file> [options]
Argument What it is
meeting required stored meeting id or meeting reference.
file required a recording: WAV as it is, MP4, M4A or other audio and video through an installed ffmpeg.
Option What it does
--model <model> installed speech model.
--models-dir <dir> shared installed speech-model directory.
--max-bytes <bytes> largest audio file to read, in bytes (default 1 GiB).
--max-duration <duration> longest audio to transcribe (default 4h).
--max-segments <n> most recognized segments (default 10000).
--max-text-bytes <bytes> most transcript text, in bytes (default 4 MiB).
--yes recognize locally and append to the existing meeting.
--dry-run preview even when --yes is present.

Returns: meeting transcription — applied, account_id, external_id, segments, media, meeting_id, created, inserted_transcripts, and 7 more; every field: zm commands info meetings transcribe

Example: zm personal meetings transcribe meeting:3/12 ./AliceExample.m4a --model parakeet-v3 --json

zm meetings evidence

read the stored transcript lines, chat or summary behind a meeting reference

zm meetings evidence <reference> [options]
Argument What it is
reference required meeting reference, such as meeting:3/15.
Option What it does
--cues <n> how many transcript lines, 1 through 1000.
--bytes <bytes> how much text, in bytes, 2 through 4194304.
--cursor <cursor> continue after this line reference from the previous page.
--after-transcript-id <id> resume root-reference revision scan, including empty revisions.
--max-read-bytes <bytes> most stored bytes to read per page (default 4 MiB).
--max-transcript-pages <n> most transcript pages to read, 1 through 1000.

Returns: meeting evidence — schema_version, kind, source, fingerprint, limits, content_bytes, coverage, items; every field: zm commands info meetings evidence

Example: zm personal meetings evidence meeting:3/15 --json

zm meetings context

what the stored meetings say about a person, through linked identities

zm meetings context <person> [options]
Argument What it is
person required stored person, such as person:7.
Option What it does
--cues <n> how many transcript lines, 1 through 1000.
--bytes <bytes> how much text, in bytes, 2 through 4194304.
--meetings <n> how many meetings, 1 through 100.
--scan-limit <n> most identity links to look at, 1 through 1000.
--since-time <time> only identity links since this ISO 8601 time.

Returns: meeting person context — person, items, limits, content_bytes, coverage; every field: zm commands info meetings context

Example: zm personal meetings context person:7 --meetings 10 --json

zm meetings task-proposal

preview the task a meeting line would become, without creating it

zm meetings task-proposal <reference> [options]
Argument What it is
reference required transcript line reference, such as meeting:3/15/20/0.
Option What it does
--type <type> question, request, mention or promise.

Returns: meeting task proposal — id, source, source_kind, group, account_id, kind, preview, revision, and 2 more; every field: zm commands info meetings task-proposal

Example: zm personal meetings task-proposal meeting:3/15/20/0 --type promise --json

zm meetings passage

read a cited cue with bounded context before and after it

zm meetings passage <reference> [options]
Argument What it is
reference required retained transcript cue reference from a search hit.
Option What it does
--before-n <n> preceding cues (default 3).
--after-n <n> following cues (default 3); at most 999 surrounding cues in total.
--bytes <bytes> cue JSON byte budget (default 65536); the anchor must fit.
--max-read-bytes <bytes> stored text byte budget (default 4194304); boundary probes read metadata only.
--expected-fingerprint <hash> reject a changed anchor using its previously read cue fingerprint.
--format <format> compact text, Markdown or structured JSON. One of: text, markdown, json. Default: text.

Returns: meeting passage — schema_version, kind, source, target_ref, target_fingerprint, fingerprint, revision, limits, and 3 more; every field: zm commands info meetings passage

Example: zm personal meetings passage meeting:1/1/1/0 --json

zm meetings embeddings

preview building meeting vectors; --yes builds one batch

Changes local data.

zm meetings embeddings [options]
Option What it does
--model <model> embedding model: an installed local one, or one served at --base-url.
--base-url <url> remote embedding service; OpenAI reads OPENAI_API_KEY.
--dims <n> vector size the remote service returns, in dimensions.
--limit <n> how many results or items in one batch, 1 through 1000.
--max-rows <n> most transcript lines to read, 1 through 100000.
--max-text-bytes <bytes> most transcript text to read, in bytes (default 4 MiB).
--after-transcript-id <id> continue splitting transcripts after this one.
--after-hash <hash> continue embedding after this passage.
--yes split transcripts and build their vectors; otherwise only preview.

Returns: meeting embedding — account_id, model, status, applied, scope, rebuild_required, saved, skipped, and 3 more; every field: zm commands info meetings embeddings

Example: zm personal meetings embeddings --model e5-small --json

zm meetings status

the last meeting ingestion pass: confirmed count, completion and checkpoint age

zm meetings status

Returns: sync progress — job, profile, done, total, remaining, state, updated_at, cursor_age_ms; every field: zm commands info meetings status

Example: zm personal meetings status --json

zm events

the owner's stored events

zm events list

list stored events

zm events list [query]
Argument What it is
query optional event id, origin or time fields that narrow the list.

Returns: event — id, event_series_id, title, description, location, starts_at, ends_at, timezone, and 13 more; every field: zm commands info events list

Example: zm personal events list --json

zm commands

every command by section; --json prints its contract, one path per call

zm commands info

one command contract, or all schemas, error reasons and explicit overrides

zm commands info [path]
Argument What it is
path optional one command path; without one, export schemas, error reasons and explicit overrides.

Example: zm commands info commands

zm store

the shared local store every WireCat tool reads

zm store info

the store file: where it is, its size, its schema against the one this tool speaks, which tools used it and how many rows it holds, by provider, chat kind, account, year (UTC) and attachment kind, over every account; changes nothing

zm store info

Returns: store info — path, exists, schema, opens, error, bytes, journal_mode, pages, and 12 more; every field: zm commands info store info

Example: zm personal store info --json

zm complete

answer shell completion requests

zm complete [words]
Argument What it is
words optional shell words ending in the partial word.

Example: zm complete -- meetings s

zm session

the saved Zoom login, separate from local data profiles

zm session status

show cached login state; remote verification is explicit

zm session status [options]
Option What it does
--check verify the session with Zoom.

Example: zm personal session status --json, zm personal session status --check --dry-run --json

zm session start

log in to Zoom through your browser

Changes remote Zoom data.

zm session start [options]
Option What it does
--browser <program> browser program.
--client-id <id> the Public Client ID of your Zoom app; saved as zoom.client_id once you sign in.
--credential-storage <storage> where the login is kept: keyring (the default, which falls back to a private file by itself when no keyring works), or file, not recommended. Default: keyring.
--answer-stdin with --json and no terminal: print each question as a needs line on stderr and read its answer, one line, from stdin.

Example: zm personal session start

zm session end

delete this profile's saved Zoom login; the stored meetings stay

Changes local data.

zm session end

Example: zm personal session end

zm upgrade

upgrade zm with the package manager that installed it; --check only looks

Changes local data.

zm upgrade [options]
Option What it does
--check say whether a newer version exists, and install nothing.

Example: zm upgrade --check

zm models

models that run on this machine

zm models list

speech and embedding models, their download status and the default of each kind

zm models list [query]
Argument What it is
query optional field:value terms that narrow the list — the fields are below.

Returns: model — id, kind, title, languages, size_bytes, downloaded, default, path, and 6 more; every field: zm commands info models list

Next usually: zm models download <model>

Example: zm models list, zm models list kind:audio

zm models download

download and verify one speech or embedding model

Changes local data.

zm models download <model> [options]
Argument What it is
model required a model id from models list.
Option What it does
--offline refuse a download that needs the network.
--accept-terms no longer needed; kept so older scripts still run.

Returns: model — id, kind, title, languages, size_bytes, downloaded, default, path, and 6 more; every field: zm commands info models download

Example: zm models download parakeet-v3

zm account

the profile's login and data on this computer

zm account remove

sign the profile out, then forget it on this computer: its settings, its saved sign-in and its files; --keep-data leaves its messages in the store, where search still finds them, --delete-data deletes them; asks first, or --yes

Changes remote Zoom data.

zm account remove [options]
Option What it does
--keep-data leave what the store holds of this account, still searchable.
--delete-data delete what the store and the tool's own files hold of this account.
--dry-run list what it would remove and keep, and change nothing.

Returns: account removal — profile, data, dry_run, account, revoked, config_file, removed, kept; every field: zm commands info account remove

Example: zm personal account remove --keep-data --dry-run, zm personal account remove --delete-data --yes

zm watch

keep fetching new meetings from Zoom, or rescanning a folder, until stopped

Changes local data.

zm watch [target] [folder] [options]
Argument What it is
target optional api or files; otherwise the configured source. One of: api, files.
folder optional with files: folder of downloaded transcripts; otherwise files.folder.
Option What it does
--since-time <time> on the first run, fetch meetings from this ISO 8601 time.
--lookback-days <days> API cursor overlap for late transcripts (0 through 31).
--pause <duration> delay after each completed receipt (1s through 24h). Default: 10000.

Returns: meeting ingestion — meetings, participants, transcript_rows, summaries, warnings, timezone; every field: zm commands info watch

Example: zm personal watch files ./downloads --pause 10s --jsonl, zm personal watch api --lookback-days 2 --dry-run --jsonl

zm recordings

cloud recording assets and verified local archives

zm recordings list

list cloud recording occurrences (last 30 days by default)

zm recordings list [query] [options]
Argument What it is
query optional field:value terms; this list filters by none yet.
Option What it does
--since-time <time> first day, YYYY-MM-DD.
--until-time <time> last day, YYYY-MM-DD (default today in --timezone).
--timezone <zone> IANA timezone of today (default the computer's).

Returns: zoom recording — id, uuid, title, started_at, assets; every field: zm commands info recordings list

Example: zm personal recordings list --since-time 2026-10-01 --until-time 2026-10-10 --dry-run --json

zm recordings show

show downloadable asset metadata without signed URLs

zm recordings show <meeting>
Argument What it is
meeting required meeting id or occurrence UUID.

Returns: zoom recording — id, uuid, title, started_at, assets; every field: zm commands info recordings show

Example: zm personal recordings show example-occurrence --dry-run --json

zm recordings download

download one recording or a date range into a verified, resumable archive

Changes local data.

zm recordings download [meeting] [options]
Argument What it is
meeting optional meeting id or occurrence UUID; omit when downloading a date range.
Option What it does
--output-dir <dir> explicit archive directory.
--since-time <time> first day of a date range, YYYY-MM-DD.
--until-time <time> last day of a date range, YYYY-MM-DD; at most ten years after the first.
--file-ids <ids> comma-separated recording file ids; only with a meeting.
--max-bytes <bytes> per-file download byte limit (default 20 GiB).
--resume resume verified partial downloads and archive progress; requires a local profile.
--no-resume disable persistent resume state for this invocation.
--checkpoint <file> date-range checkpoint file; implies resume.
--recheck-since-time <time> look again from this day for late recordings in a date range.
--max-pages <pages> date-range catalog page limit from 1 through 10000 (default 1000).
--max-entries <n> date-range occurrence limit from 1 through 10000 (default 10000).

Returns: zoom archive — uuid, manifest_path, downloaded, skipped, unavailable, window, meeting, archive, and 2 more; every field: zm commands info recordings download

Example: zm personal recordings download example-occurrence --output-dir ./zoom-archive --dry-run --json, zm personal recordings download --since-time 2026-01-01 --until-time 2026-10-10 --output-dir ./zoom-archive --dry-run --jsonl

zm recordings archive

inspect archives without Zoom or a database

zm recordings archive transcript

read the transcripts of a local archive, checking each file first

zm recordings archive transcript <directory> [options]
Argument What it is
directory required archive folder.
Option What it does
--uuid <uuid> select an occurrence UUID.
--file-ids <ids> select comma-separated transcript asset ids.
--max-bytes <bytes> largest transcript file to read, in bytes (default 8 MiB).

Returns: zoom archive transcript — meeting, asset, cues; every field: zm commands info recordings archive transcript

Example: zm personal recordings archive transcript ./zoom-archive --uuid example-occurrence --json

zm recordings archive show

show what a local archive holds, from its transfer records

zm recordings archive show <directory>
Argument What it is
directory required archive folder.

Returns: zoom archive manifest — version, meeting, files; every field: zm commands info recordings archive show

Example: zm personal recordings archive show ./zoom-archive --json

zm recordings archive check

check the size and checksum of every file in a local archive

zm recordings archive check <directory>
Argument What it is
directory required archive folder.

Returns: zoom archive check — manifest_path, id, path, valid; every field: zm commands info recordings archive check

Example: zm personal recordings archive check ./zoom-archive --json

zm recordings archive locks

list the locks on a local archive: active, stale or another machine's

zm recordings archive locks <directory>
Argument What it is
directory required archive folder.

Returns: zoom archive lock — name, status, pid, host; every field: zm commands info recordings archive locks

Example: zm personal recordings archive locks ./zoom-archive --json

zm recordings archive recover

preview removing stale locks; --yes removes them

Changes local data.

zm recordings archive recover <directory> [options]
Argument What it is
directory required archive folder.
Option What it does
--yes remove recognized locks left by a dead local process.
--dry-run preview only, including when --yes is present.

Returns: zoom archive recovery — locks, removed; every field: zm commands info recordings archive recover

Example: zm personal recordings archive recover ./zoom-archive --dry-run --json

zm recordings archive import

check an archive's transcripts, then add them to the stored meetings

Changes local data.

zm recordings archive import <directory> [options]
Argument What it is
directory required archive folder.
Option What it does
--uuid <uuid> select an occurrence UUID.
--file-ids <ids> select comma-separated transcript asset ids.
--max-bytes <bytes> largest transcript file, in bytes (default 8 MiB).
--max-total-bytes <bytes> most transcript bytes in all, checked before any write (default 64 MiB).

Returns: zoom archive import — uuid, asset_ids, status, meeting_id, created, inserted_transcripts, replayed_transcripts, superseded_transcripts, and 1 more; every field: zm commands info recordings archive import

Example: zm personal recordings archive import ./archive --jsonl

zm recordings schedule

local archive jobs and Linux user systemd timers

zm recordings schedule create

preview a daily archive job; --yes installs its timer

Changes local data.

zm recordings schedule create [options]
Option What it does
--name <name> local job name.
--output-dir <dir> absolute archive directory.
--since-time <time> first day to archive, YYYY-MM-DD.
--at <hh:mm> daily time.
--timezone <zone> IANA timezone for the timer.
--lookback-days <days> revisit the latest zero through 31 UTC days (default seven).
--max-bytes <bytes> maximum bytes per recording file (default 20 GiB).
--yes install and enable the captured timer.
--dry-run preview even when --yes is present.

Returns: zoom archive schedule change — applied, backend, job, timer_state, units; every field: zm commands info recordings schedule create

Example: zm personal recordings schedule create --name daily --output-dir /home/AliceExample/archive --since-time 2026-01-01 --at 03:15 --timezone UTC --json

zm recordings schedule list

list saved local jobs; live timer state is not checked

zm recordings schedule list [query]
Argument What it is
query optional field:value terms; this list filters by none yet.

Returns: zoom archive job — version, schedule, capture, state; every field: zm commands info recordings schedule list

Example: zm personal recordings schedule list --offline --json

zm recordings schedule show

show one saved local job and its captured settings

zm recordings schedule show <job>
Argument What it is
job required saved job id.

Returns: zoom archive job — version, schedule, capture, state; every field: zm commands info recordings schedule show

Example: zm personal recordings schedule show example-job --offline --json

zm recordings schedule run

run the captured job once with durable archive progress

Changes local data.

zm recordings schedule run <job>
Argument What it is
job required saved job id.

Returns: zoom archive receipt — window, meeting, archive, availability, checkpoint_revision; every field: zm commands info recordings schedule run

Example: zm personal recordings schedule run example-job --dry-run --jsonl

zm recordings schedule disable

preview disabling a saved timer; --yes disables it without deleting files

Changes local data.

zm recordings schedule disable <job> [options]
Argument What it is
job required saved job id.
Option What it does
--yes disable and stop the timer.
--dry-run preview even when --yes is present.

Returns: zoom archive schedule change — applied, backend, job, timer_state, units; every field: zm commands info recordings schedule disable

Example: zm personal recordings schedule disable example-job --json

zm recordings status

confirmed archived assets and pending discovery in one archive checkpoint

zm recordings status <directory> [options]
Argument What it is
directory required archive folder.
Option What it does
--checkpoint <file> an explicit archive checkpoint path.

Returns: sync progress — job, profile, done, total, remaining, state, updated_at, cursor_age_ms; every field: zm commands info recordings status

Example: zm personal recordings status ./zoom-archive --json

zm contacts

your Zoom contacts, read live: company colleagues and external contacts

zm contacts list

list your company and external Zoom contacts

zm contacts list [query]
Argument What it is
query optional field:value terms that narrow the list — the fields are below.

Returns: contact — ref, id, name, username, description, last_messaged_at, alias, display_name, and 15 more; every field: zm commands info contacts list

Next usually: zm meetings context <person>

Example: zm personal contacts list --json, zm personal contacts list type:external name:alice

zm contacts show

one contact with their phone numbers

zm contacts show <contact>
Argument What it is
contact required their member id from contacts list, Zoom user id or email.

Returns: contact — ref, id, name, username, description, last_messaged_at, alias, display_name, and 15 more; every field: zm commands info contacts show

Next usually: zm meetings context <person>

Example: zm personal contacts show alice@example.com --json

zm zoom

Zoom's own live search: chat, docs and meetings zm does not archive

zm zoom search

search your Zoom workspace live; without words, list recent records

zm zoom search [query] [options]
Argument What it is
query optional words Zoom matches in the content; it does not widen word forms.
Option What it does
--only <resources> comma list of meetings, calendar, chat, channels, docs, notes, calls, sms, voicemail (default Zoom's own choice).
--limit <n> Zoom results per resource, 1 through 100 (default 10); one chat result holds its matching messages.
--cursor <cursor> continue from the previous page's page.next_cursor, with the same query and --only.

Returns: zoom search hit — source, id, title, time, text, url, people, container_id, and 1 more; every field: zm commands info zoom search

Example: zm personal zoom search "budget" --only chat,docs --json, zm personal zoom search --only meetings --limit 50

zm skill

the instructions an agent is given for this tool

zm skill show

print SKILL.md — zm skill install puts it where Claude Code, Codex, Gemini CLI and OpenClaw look for it, and where Hermes does on request; --for prints the copy already there

zm skill show [name] [options]
Argument What it is
name optional a part of this skill, or a skill shipped for a task: zoom-live, setup, search, prepare, people-and-projects, archive.
Option What it does
--for <agents> print the copy installed for these agents instead of this version's. One of: claude, agents.

Example: zm skill show --json

zm skill install

write SKILL.md to ~/.claude/skills/zoom-cli/ (Claude Code) and ~/.agents/skills/zoom-cli/ (Codex, Gemini CLI, OpenClaw); --for hermes also has Hermes read ~/.agents/skills, which changes Hermes's own settings, so all leaves it out

Changes local data.

zm skill install [options]
Option What it does
--for <agents> which agents to install for. One of: claude, agents, hermes, all. Default: all.

Example: zm skill install --for agents --json

zm skill status

for each place zm skill install writes to, and for the copy Hermes reads: the path, the version installed there, and whether it is current, stale, missing or unversioned

zm skill status

Example: zm skill status, zm skill status --json

zm search

opt-in shared store search; meetings by default

zm search all

search meetings and, with --only, messages, mail and notes in the local store

zm search all <query> [options]
Argument What it is
query required words to find; a word also matches longer words it starts.
Option What it does
--only <resources> comma-separated meetings,messages,mail,notes (default meetings).
--limit <n> how many results, 1 through 1000 (default 100). Default: 100.
--page <n> which page, starting at 1.
--exact request exact matching; unsupported meeting queries are reported as skipped.

Returns: hit — kind, ref, provider, account, title, timestamp, excerpt, url, and 18 more; every field: zm commands info search all

Example: zm personal search all "example plan" --json

zm search meetings

find words in stored meetings, or by meaning with --model

Changes local data.

zm search meetings <query> [options]
Argument What it is
query required words, OR, (groups), -word, "a phrase", "a b"~N (within N lines), exact:word, from:"Name" (who said the line); date>=30d, project:, tag:, topic: choose meetings; a word also matches its other forms and spellings speech recognition garbled.
Option What it does
--meeting-id <id> only this stored meeting.
--identity-id <id> meetings with this participant identity; does not say who spoke a hit.
--person-id <id> meetings with a participant linked to this person.
--event-id <id> only meetings linked to this stored event.
--series-id <id> only meetings in this stored series.
--project <project> meetings confirmed in this project, by name, key or alias.
--tag <name> meetings with this tag, by name or alias.
--topic-name <name> meetings with this topic, by name or alias.
--since-time <time> meetings starting at or after this ISO 8601 date or time, or 30m, 2h, 7d ago.
--until-time <time> meetings starting at or before this ISO 8601 date or time; a date alone covers that whole local day.
--limit <n> how many hits, 1 through 1000; the profile's defaults.limit when absent.
--cursor <cursor> continue from the previous page's page.next_cursor, with the same query and filters.
--max-read-bytes <bytes> most stored bytes to read, 1 through 67108864 (default 4194304).
--model <model> rank words and meaning together with this embedding model, building missing vectors first.
--base-url <url> remote embedding service for --model; OpenAI reads OPENAI_API_KEY.
--dims <n> vector size the remote service returns, in dimensions.
--max-chunks <n> by meaning: most stored passages to compare (default 10000).
--max-rows <n> by meaning: most transcript lines to read, 1 through 100000.
--max-text-bytes <bytes> by meaning: most transcript text to read (default 4 MiB).

Returns: meeting hit — ref, meeting, scope, id, excerpt, offset_ms, language, line, and 8 more; every field: zm commands info search meetings

Next usually: zm meetings show <meeting>, zm meetings transcript <meeting>, zm meetings summary <meeting>

Example: zm personal search meetings "example plan" --since-time 2026-09-01 --json, zm personal search meetings "example plan" --model e5-small --json

zm search recordings

find words in the transcripts of a downloaded archive folder, before any import

zm search recordings <directory> <query> [options]
Argument What it is
directory required archive folder.
query required words to find as one phrase, ignoring case.
Option What it does
--limit <n> how many hits, 1 through 1000 (default 100).

Returns: hit — kind, ref, provider, account, title, timestamp, excerpt, url, and 18 more; every field: zm commands info search recordings

Example: zm personal search recordings ./zoom-archive "example plan" --limit 100 --offline --json

zm stats

numbers about the stored meetings

zm stats meetings

talk time by speaker and week; meetings and minutes by day, week or series

zm stats meetings show

meetings, lines, words and talk time by week, with the top speakers

zm stats meetings show [options]
Option What it does
--since-time <time> meetings starting at or after this date or time (ISO 8601, or 30d ago).
--until-time <time> meetings starting at or before this date or time; a date alone is that whole day.
--timezone <zone> the IANA zone dates and weeks are read in (default: this computer's).
--event-id <id> only meetings linked to this stored event.
--project <project> meetings confirmed in this project, by name, key or alias.
--topic-name <name> meetings with this topic label, by name or alias.
--max-rows <n> most transcript lines to read (default and at most 500000).

Returns: meeting stats — time_zone, meetings, lines, words, talk_ms, weeks, top_speakers; every field: zm commands info stats meetings show

Example: zm personal stats meetings show --since-time 30d --json

zm stats meetings top

speakers by talk time: lines, words, meetings and their share of the talk

zm stats meetings top [options]
Option What it does
--since-time <time> meetings starting at or after this date or time (ISO 8601, or 30d ago).
--until-time <time> meetings starting at or before this date or time; a date alone is that whole day.
--timezone <zone> the IANA zone dates and weeks are read in (default: this computer's).
--event-id <id> only meetings linked to this stored event.
--project <project> meetings confirmed in this project, by name, key or alias.
--topic-name <name> meetings with this topic label, by name or alias.
--max-rows <n> most transcript lines to read (default and at most 500000).
--limit <n> how many speakers (default 20).

Returns: meeting speaker — key, name, person_id, identity_id, meetings, lines, words, talk_ms, and 1 more; every field: zm commands info stats meetings top

Example: zm personal stats meetings top --project RES --limit 5 --json

zm stats meetings history

meetings, minutes and meetings with a transcript, per day, week or series; reads meetings only

zm stats meetings history [options]
Option What it does
--since-time <time> meetings starting at or after this date or time (ISO 8601, or 30d ago).
--until-time <time> meetings starting at or before this date or time; a date alone is that whole day.
--timezone <zone> the IANA zone dates and weeks are read in (default: this computer's).
--event-id <id> only meetings linked to this stored event.
--project <project> meetings confirmed in this project, by name, key or alias.
--topic-name <name> meetings with this topic label, by name or alias.
--by <group> one row per day, week, series.
--limit <n> the newest days or weeks, or the busiest series (default 100).

Returns: meeting counts — by, meetings, minutes, mean_minutes, transcripts, rows, has_more; every field: zm commands info stats meetings history

Example: zm personal stats meetings history --by week --since-time 90d --json

zm review

meetings since a point: summary, next steps and who talked how long

zm review [options]
Option What it does
--since-time <time> meetings starting at or after this date or time (ISO 8601, or 30d ago).
--until-time <time> meetings starting at or before this date or time; a date alone is that whole day.
--timezone <zone> the IANA zone dates and weeks are read in (default: this computer's).
--event-id <id> only meetings linked to this stored event.
--project <project> meetings confirmed in this project, by name, key or alias.
--topic-name <name> meetings with this topic label, by name or alias.
--limit <n> most meetings, newest first (default 50, at most 200).

Returns: meeting review — ref, id, title, started_at, duration_ms, participants_count, projects, transcript, and 2 more; every field: zm commands info review

Example: zm personal review --json, zm personal review --since-time 30d --project RES --json

zm mentions

what the stored meetings point to: links shared in their chat

zm mentions list

links shared in meeting chat, most shared first, by whom and where

zm mentions list [query] [options]
Argument What it is
query optional field:value terms that narrow the list — the fields are below.
Option What it does
--timezone <zone> the IANA zone dates and weeks are read in (default: this computer's).
--limit <n> how many links (default 100).
--max-messages <n> most chat messages to read (default and at most 100000).

Returns: meeting link — url, domain, document_id, count, first_at, last_at, senders, meetings; every field: zm commands info mentions list

Example: zm personal mentions list started_at>=30d --json, zm personal mentions list project:RES --json

zm searches

saved meeting searches: a query and its options under a name

zm searches create

save a search meetings query and its options without running it

Changes local data.

zm searches create <name> <query> [options]
Argument What it is
name required up to 64 letters a–z, digits and hyphens, not only digits.
query required the query, as search meetings takes it.
Option What it does
--replace overwrite a saved search of the same name.
--event-id <id> only meetings linked to this stored event.
--project <project> meetings confirmed in this project, by name, key or alias.
--topic-name <name> meetings with this topic, by name or alias.
--since-time <time> meetings starting at or after this ISO 8601 date or time, or 30m, 2h, 7d ago.
--until-time <time> meetings starting at or before this ISO 8601 date or time; a date alone covers that whole local day.
--limit <n> how many hits, 1 through 1000; the profile's defaults.limit when absent.
--max-read-bytes <bytes> most stored bytes to read, 1 through 67108864 (default 4194304).
--model <model> rank words and meaning together with this embedding model, building missing vectors first.
--base-url <url> remote embedding service for --model; OpenAI reads OPENAI_API_KEY.
--dims <n> vector size the remote service returns, in dimensions.
--max-chunks <n> by meaning: most stored passages to compare (default 10000).
--max-rows <n> by meaning: most transcript lines to read, 1 through 100000.
--max-text-bytes <bytes> by meaning: most transcript text to read (default 4 MiB).

Returns: saved search — id, name, command, params, language, version, fields_version, created_at, and 2 more; every field: zm commands info searches create

Example: zm personal searches create weekly-plan "example plan date>=7d" --project RES --json

zm searches run

run a saved search; answers as search meetings does

Changes local data.

zm searches run <name>
Argument What it is
name required a saved search's name.

Returns: meeting hit — ref, meeting, scope, id, excerpt, offset_ms, language, line, and 8 more; every field: zm commands info searches run

Next usually: zm meetings show <meeting>, zm meetings transcript <meeting>, zm meetings summary <meeting>

Example: zm personal searches run weekly-plan --json

zm searches list

the saved meeting searches, by name

zm searches list [query]
Argument What it is
query optional field:value terms; this list filters by none yet.

Returns: saved search — id, name, command, params, language, version, fields_version, created_at, and 2 more; every field: zm commands info searches list

Example: zm personal searches list --json

zm searches show

one saved search: its query, options and how often it ran

zm searches show <name>
Argument What it is
name required a saved search's name.

Returns: saved search — id, name, command, params, language, version, fields_version, created_at, and 2 more; every field: zm commands info searches show

Example: zm personal searches show weekly-plan

zm searches delete

delete a saved search

Changes local data.

zm searches delete <name>
Argument What it is
name required a saved search's name.

Returns: saved search — id, name, command, params, language, version, fields_version, created_at, and 2 more; every field: zm commands info searches delete

Example: zm personal searches delete weekly-plan --dry-run

zm batch

run bounded stored meeting reads: search meetings and meetings passage

Changes local data.

zm batch <file> [options]
Argument What it is
file required JSON request file, or - for stdin.
Option What it does
--max-read-bytes <bytes> aggregate stored-text read ceilings, including one 4 MiB index pass; at most 67108864.
--format <format> text, markdown or json. One of: text, markdown, json. Default: text.

Returns: meeting batch result — id, command, status, answer, complete, error; every field: zm commands info batch

Example: zm personal batch requests.json --json

zm guide

scenarios: the commands for a common task, in order, and why each

zm guide [name]
Argument What it is
name optional one scenario; without it, the list.

Example: zm guide, zm guide archive-recordings --json

Exit codes

Use the shared exit code rather than matching diagnostic text.

Code When
0 it worked
2 validation_error
3 configuration_error
4 authentication_error
5 permission_error
6 not_found
7 confirmation_required
8 rate_limited
9 timeout
10 network_error
11 provider_error
12 provider_unavailable
13 invalid_response
14 outcome_unknown
15 conflict
130 cancelled
1 anything else

Remote meeting changes preview by default. Exit code 14 reports an uncertain dispatched outcome; inspect the remote state before retrying. Foreground watches require JSONL when output is piped.

Clone this wiki locally