Tool reference

These are the tools agents get. The descriptions below are the ones the server sends in tools/list, generated from the server’s own registry. Card references are the ones shown on the board, like YK-12.

project is always optional. Leave it out and a tool uses the project the session last worked in: the one it last claimed or filed a card in, or named with introduce’s project. A session that hasn’t worked anywhere yet gets the oldest active project the token can reach (find_cards searches all of them instead). Card tools find the project from the card’s ref; they also take a project, and refuse a card from another project.

An unknown argument is refused with the argument it most likely meant and the ones the tool takes, like Unknown argument "mesage". Did you mean "message"? Accepted arguments: card, message, percent, step.

What a missing swimlane means depends on the tool:

• add_card, add_lane and add_flow_rule use the first swimlane.
• get_next_card and get_flow_rules look in every swimlane; name one to narrow it down.
• find_cards treats it as a filter.
• move_card keeps the card in its own swimlane.
• update_lane and remove_lane only need it when two swimlanes have a lane of that name.

What a token sees depends on its access. A read-only token gets the read tools. A read-and-write token also gets everything for working cards, planning, labels and the archive. Deleting cards and changing lanes, swimlanes and flow rules stay with people unless the token was created with Delete and restore cards or Manage lanes, swimlanes and flows turned on; without them those tools aren’t listed at all.

The server also serves MCP prompts (work_on_card and plan_project, rendered from the project’s own prompt templates) and resources (yokka://board/{project} and yokka://card/{ref}).

introduce

Introduce yourself · needs a write token

Tell the board who you are, once at the start of a session (and again if the human renames you). Use the name the human knows this session by in their CLI or app: its session or thread title, its terminal tab, or its worktree or branch, like "auth-refactor". The board shows it on the cards you hold and in its list of agents, so a human running many agents can tell which is which, and can ask any agent about you by that name. Leave the name out and the board picks one. Names are unique among live agents: when yours is taken you get a numbered one. Tell the human the name you got, and title your session with it if your client lets you. When the answer gives you an `as` name, other chats share your connection: pass it as `as` to claim_card from then on, so the board knows the cards are yours. Pass `project` to name the project you'll work in: tools then use it when you leave `project` out.

Argument	Type	Description
name	string	Optional: your name on the board, like this session's title, terminal tab or branch ("auth-refactor"). Leave it out and the board picks one. (max 40 chars)
where	string	Optional one line on where you run, like "kanban-app · feat/auth · work laptop". Shown with your name. (max 120 chars)
as	string	Optional: the `as` name you already have, to rename yourself to `name`. (max 40 chars)
project	string	Optional project slug, name or card prefix to work in: tools use it when you leave out `project`, until you claim or file a card in another one. (max 80 chars)

list_projects

List projects · read-only, available to read-only tokens

List the projects in this workspace with their swimlanes, each swimlane's lanes and how many cards each lane holds (a done lane counts its latest cards; `more` says it holds more). Start here if you don't know which project to work in. Archived projects are read-only and left out unless include_archived.

Argument	Type	Description
include_archived	boolean	Also list archived (read-only) projects.

get_board

Read a board · read-only, available to read-only tokens

Read one project's board: every swimlane, its lanes and their cards (ref, title, who holds it, whether it needs the human, which cards it is blocked by). Swimlanes split the board by kind of work (Features, Bugs…) and each has its own lanes. A swimlane may also have a Backlog: parked work that isn't ready, which get_next_card skips. Done lanes (Shipped, Fixed…) show only their latest cards; find older ones with find_cards. Also lists the open epics (larger pieces of work that group cards across swimlanes) with their progress, each card's epic, and the project's labels. Use it to get oriented before picking work or planning.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)

get_card

Read a card · read-only, available to read-only tokens

Read a card in full: its brief (what, why, done-when), swimlane, lane, labels, epic, holder, checklist (items the human wrote are acceptance criteria), the cards it is blocked by and the cards it blocks, attached files (with download links: the human may attach screenshots or mockups), linked GitHub pull requests (with their checks), branches and commits, and recent activity. Read the brief before you start working on a card. Its cursor lets get_card_activity show only what's new. An archived card reads the same, but is read-only.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)

get_card_activity

Read a card's activity · read-only, available to read-only tokens

Read a card's activity oldest first: comments, questions, progress, moves and edits, with who did each. After request_input, pass the cursor it returned as `after` to see only what happened since, and look for a comment from a person: that's the human's answer. To wait for it, add wait_seconds (up to 50): the call answers as soon as a person or another agent adds something after the cursor, instead of you polling. Without `after`, returns the latest entries. Pass the returned cursor next time to keep reading forward.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
after	string	Optional cursor from request_input, comment_card, get_card or an earlier call. (max 80 chars)
limit	integer	Optional, 1–100 entries (default 30). (1–100)
wait_seconds	integer	Optional, with `after`: wait up to this many seconds (1–50) for a person or another agent to add something, and answer as soon as they do. Counts as one call. (1–50)

find_cards

Find cards · read-only, available to read-only tokens

Search cards by text in the title or brief, label, epic, swimlane or lane, across every active project unless you name one (name an archived project to search it). held_by_me lists the cards you hold (useful after a restart); held_by lists another agent's, by the name it introduced itself with, when the human asks about it ("what is auth-refactor doing?"). needs_human lists cards waiting on the human. in_trash searches deleted cards instead, and archived the cards put away off the board. Backlogs are included. Results come a page at a time: when `more` is true, pass the returned `cursor` with the same filters for the next page (a page can be short when a search reads many cards).

Argument	Type	Description
project	string	Optional project slug, name or card prefix. Omit to search every active project. (max 80 chars)
text	string	Words to look for in the title or brief. (max 200 chars)
label	string	Only cards with this label. (max 24 chars)
epic	string	Only cards in this epic. (max 60 chars)
swimlane	string	Only cards in this swimlane. (max 40 chars)
lane	string	Only cards in lanes with this name, like "Ready". (max 40 chars)
held_by_me	boolean	Only cards you hold, and your helpers' (agents that claimed with `as`).
as	string	Optional: the name you claim with (`as` on claim_card), so held_by_me lists only your cards. (max 40 chars)
held_by	string	Only cards held by this agent, including ones it finished: its name ("auth-refactor") or its client's ("codex"). Agents seen today only. (max 40 chars)
needs_human	boolean	Only cards flagged for the human.
in_trash	boolean	Search deleted cards instead of live ones.
archived	boolean	Search archived cards instead of live ones.
limit	integer	Optional, 1–100 cards a page (default 30). (1–100)
cursor	string	Optional: the cursor a previous page returned, to read the next one. (max 1000 chars)

get_next_card

Find the next card · read-only, available to read-only tokens

Find the top unclaimed card waiting to be picked up: in each swimlane, the lane just before the one claimed cards move to (usually Ready or Triage). Swimlanes are searched top to bottom unless you name one. Backlogs are never searched, and cards blocked by work that isn't done yet are skipped. Name an epic to only pick up that epic's cards. It only looks: claim the card with claim_card before starting, or call claim_card without a card to find and claim the next one in one step, so no other agent takes it in between.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane	string	Swimlane name, like "Bugs". Optional; leave it out to look in every swimlane. (max 40 chars)
epic	string	Epic name, like "Checkout v2". get_board lists the open epics. (max 60 chars)

claim_card

Claim a card · needs a write token

Claim a card when you start working on it, not a batch ahead of time. The board moves it (usually to Working) and shows you as its agent. Refuses a card another agent holds, including another running session on your token. Leave out `card` to take the next card waiting to be picked up, exactly as get_next_card would find it (same project, swimlane and epic filters), in one step: cards other agents hold are skipped, and no other agent can take it in between. A card blocked by other cards (they aren't done yet) is refused with what it waits on; pass force: true only if the human asked you to work on it anyway. A claim lapses once its holder has been quiet for the project's stale time (30 minutes by default); then any agent may claim the card. It doesn't lapse while a question you asked with request_input waits on the human. If you are one of several agents on one connection (a subagent, or a helper another agent started), pass `as` with your own short name, like "setup-copy": the board shows you as your own agent under the one that started you, and every tool you call on this card acts as you.

Argument	Type	Description
card	string	The card reference shown on the board, like "YK-12". Optional: leave it out to claim the next card waiting. (max 80 chars)
as	string	Optional: your own name, when you share a connection with other agents (a subagent), like "setup-copy". Use the same one for every card you claim. (max 40 chars)
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane	string	Without a card: only look in this swimlane, like "Bugs". Leave it out to look in every swimlane. (max 40 chars)
epic	string	Without a card: only take a card of this epic, like "Checkout v2". (max 60 chars)
note	string	Optional one-line note on your plan. (max 280 chars)
launch	string	The launch code from your start prompt (like "r_7f3k2a"), when you were given one. It links your session to the run the person started. (max 20 chars)
force	boolean	Optional: claim the card even though it is blocked by cards that aren't done. Only when the human asked for it.

report_progress

Report progress · needs a write token

Report progress on a card you hold, at real milestones only. The human watches the board live, so keep each message to one short line ("Added the Stripe session endpoint"). Say when a long step starts (tests, a build) and how it ended, and have subagents working on the card report here too. Never report work you haven't done.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
message required	string	One line on what just happened. (max 280 chars)
percent	number	Optional rough completion, 0–100 (rounded to a whole number). Leave it out on a card with a checklist: its bar follows the items. (0–100)
step	string	Optional short step name, like "Tests". (max 60 chars)

stop_check

Check your cards before you stop · read-only, available to read-only tokens

Call it before you end your turn. It lists the cards you hold that the board shows as in progress although you haven't said anything about them for 10 minutes or more: not done, not released and not waiting on the human. Answers `{}` when there are none. Otherwise it answers with `{"decision": "block", "reason": ...}` naming each card: complete, release, request_input or report_progress on it before you stop. The reply is Claude Code hook output, so a Stop hook can call this tool directly.

Argument	Type	Description
stop_hook_active	boolean	Claude Code's stop_hook_active: true once a Stop hook already sent you back this turn, and then this check lets the turn end rather than blocking again. "true" and "false" are accepted as well.

set_checklist

Write a card's checklist · needs a write token

Write your plan for a card as a checklist the human can follow on the board: 3–10 concrete steps, in order, each one short line. Call it again to revise the plan; pass the full list each time. Items whose text is unchanged keep their number and ticked state; your other items are dropped. Items the human wrote (acceptance criteria) always stay: work through them, you can't reword or remove them. Call it before you change anything; only a one-line fix may skip it. Refuses a card another agent holds.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
items required	string[]	The whole checklist in order, like ["Add the endpoint", "Write tests", "Update the docs"]. (up to 50)

check_items

Tick checklist items · needs a write token

Tick checklist items off as you finish them, by the numbers get_card and set_checklist show (#3 is 3). The board's progress bar follows the checklist, so tick each item when it's really done, not ahead of time. Pass done: false to untick. Refuses a card another agent holds.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
items required	integer[]	Item numbers, like [1, 2]. (up to 50)
done	boolean	Optional; false unticks them. Defaults to true.

attach_file

Attach proof · needs a write token

Attach proof to a card you hold: a screenshot of the change, a screen recording, a test report or a log, so the human can see it worked without running it. Returns a single-use upload link: PUT the file's raw bytes to it (the reply shows the curl command) within 15 minutes and it appears on the card. Images, MP4/WebM/MOV videos, PDFs and text files, up to 20 MB each (less on some plans: the reply gives the cap). Prefer a cropped screenshot or a short, trimmed recording. Attach before complete_card. If you can't reach the network (a sandbox) and a Yokka runner started you, pass `path` instead and the runner uploads the file from your working folder.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
filename required	string	The file's name, like "checkout-after.png". Its extension sets the type. (max 200 chars)
caption	string	Optional one line on what it shows ("Checkout page with the new totals"). (max 280 chars)
content_type	string	Optional MIME type, like "image/png", when the extension doesn't say. (max 100 chars)
path	string	Optional, runner runs only: the file's path relative to your working folder (screenshots/after.png). The runner uploads it; you don't PUT anything. (max 500 chars)

request_input

Ask the human · needs a write token

Ask the human a question when you're blocked on a decision only they can make. The card is flagged "your turn" (and usually moves to Needs you). Ask one clear question, then wait for their reply with get_card_activity, passing the returned cursor as `after` and wait_seconds so it answers as soon as they reply, before continuing. When the human should pick between a few choices, pass them as `options` so they can answer with one tap (on their phone too); for visual choices (designs, screenshots, charts), attach each with attach_file first and name the files in `option_files`, in the same order. Their reply says which option they picked (`picked`, from 1), or has no `picked` when they chose Other and wrote their own answer. If they answer you somewhere else instead (like your chat), note their answer with comment_card and `answers_question` before carrying on: that closes the question on the board.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
question required	string	One clear question. (max 4000 chars)
options	string[]	Optional: 2 to 6 short choices, like ["Rounded", "Geometric", "Wordmark only"]. (up to 6)
option_files	string[]	Optional: for each option in order, the file name of an attachment on this card to show with it ("logo-a.png"), or "-" for none. (up to 6)

comment_card

Comment on a card · needs a write token

Leave a comment on a card: answer something the human wrote, note a decision, or leave context for whoever picks it up next. Unlike request_input it doesn't flag the card for the human or move it. With `answers_question`, it records the answer to your open question that the human gave you elsewhere (like your chat) and takes the flag off. Use report_progress for milestones on work you hold.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
message required	string	The comment. Markdown is fine. (max 4000 chars)
answers_question	boolean	Optional: true when the comment is the human's answer to your open request_input question, given outside the board. Closes the question, so the card stops waiting on them.

complete_card

Complete a card · needs a write token

Mark a card you hold as done, with a one or two sentence summary of what changed. The board moves it on (usually to Review, for the human to check). Add links to pull requests or docs if you have them.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
summary required	string	What changed, in one or two sentences. (max 4000 chars)
links	string[]	Optional http(s) URLs (pull request, preview, docs). Anything else is left out. (up to 10)

release_card

Release a card · needs a write token

Give back a card you hold when you can't or shouldn't finish it. It is unassigned and usually returns to Ready. Say why in one line. Refuses a card you don't hold.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
reason	string	Why you're letting go. (max 280 chars)

report_usage

Report usage · needs a write token

Report what your work on a card cost, when your app shows it: tokens, cost in US dollars, model and active time. Send your running totals for this card so far, not just the latest step: each report replaces your last one for the card, so report as often as you like (before complete_card is a good time) and nothing is counted twice. Optional, and skip any number you don't know; runs a runner started report their usage on their own.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
model	string	Optional: the model you ran on, like "claude-opus-4-5" or "gpt-5-codex". (max 80 chars)
input_tokens	integer	Optional: input tokens, not counting cache reads. (0–1000000000000)
output_tokens	integer	Optional: output tokens, thinking included. (0–1000000000000)
cache_read_tokens	integer	Optional: input tokens read from the prompt cache. (0–1000000000000)
cache_write_tokens	integer	Optional: input tokens written to the prompt cache. (0–1000000000000)
cost_usd	number	Optional: what it cost in US dollars, as your app reports it (like 1.42). (0–1000000)
duration_ms	integer	Optional: how long you were actively working on it, in milliseconds. (0–2592000000)

add_card

Add a card · needs a write token

File a new card: for work your human asks for that isn't on the board yet (claim it with claim_card before you start), for follow-up work you discover, or when planning. Give it a clear title and a brief covering what, why and done-when, plus optional labels (new ones join the project). Goes to the first lane of the first swimlane unless you name a swimlane and/or lane. Name the lane "Backlog" to park an idea that isn't ready, on swimlanes that have a backlog.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
title required	string	Short, specific title. (max 200 chars)
brief	string	Markdown: what, why, done-when. (max 20000 chars)
swimlane	string	Swimlane name, like "Bugs". Optional; defaults to the first swimlane. (max 40 chars)
lane	string	Lane name within the swimlane, or "Backlog". Defaults to the swimlane's first lane. (max 40 chars)
labels	string[]	Optional labels, like "bug" or "frontend". Reuse existing labels where they fit. (up to 8)
epic	string	Optional open epic to file it under, by name. Create one first with add_epic if it doesn't exist. (max 60 chars)
checklist	string[]	Optional checklist: the steps, or what done means, one short line each. Whoever works the card ticks them off. (up to 50)
blocked_by	string[]	Optional cards of the same project this one waits on, like ["YK-3"]. Until they reach a done lane, get_next_card skips it and claim_card refuses it. (up to 20)

update_card

Edit a card · needs a write token

Edit a card's title, brief, labels, epic or the cards it waits on: sharpen a brief you filed, fix labels, file it under an epic, or set what it's blocked by. `labels` replaces the whole list; add_labels and remove_labels change only those, and blocked_by, add_blocked_by and remove_blocked_by work the same way. Refuses a card another agent holds.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
title	string	New title. (max 200 chars)
brief	string	New brief (markdown), replacing the old one. (max 20000 chars)
labels	string[]	The card's full new label list. (up to 8)
add_labels	string[]	Labels to add. (up to 8)
remove_labels	string[]	Labels to take off. (up to 8)
epic	string	Open epic to move it into, by name. Pass "" to take it out of its epic. (max 60 chars)
blocked_by	string[]	The full list of cards of the same project it waits on, by ref. Pass [] to clear it. (up to 20)
add_blocked_by	string[]	Cards it should also wait on, like ["YK-3"]. (up to 20)
remove_blocked_by	string[]	Cards it should stop waiting on. (up to 20)

move_card

Move a card · needs a write token

Move a card to another lane, like a person dragging it: promote it out of the Backlog, park it, or send it to another swimlane. Agent-event flow rules don't run, but an auto-start rule on the target lane may queue a run for it. Don't use it to claim, ship or release work: claim_card, complete_card and release_card do that. Refuses a card another agent holds.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)
lane required	string	Target lane, like "Ready" or "Backlog". (max 40 chars)
swimlane	string	Target swimlane. Defaults to the card's own swimlane. (max 40 chars)
position	string	Top (default) or bottom of the lane. (one of top, bottom)

add_epic

Add an epic · needs a write token

Create an epic: a larger piece of work (a feature, a migration) whose cards can sit in any swimlane. Use it when planning something that needs several cards, then file them with add_card and its epic argument. Check get_board first so you don't duplicate an existing epic.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
name required	string	Short name, like "Checkout v2". (max 60 chars)
brief	string	Optional markdown: the goal and what done means. (max 4000 chars)

update_epic

Edit an epic · needs a write token

Rename an epic, rewrite its brief, or close it when its work has shipped (closed epics keep their cards but take no new ones). closed: false reopens it.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
epic required	string	The epic to change, by name. (max 60 chars)
name	string	New name. (max 60 chars)
brief	string	New brief (markdown). (max 4000 chars)
closed	boolean	true closes the epic, false reopens it.

add_label

Add a label · needs a write token

Add a label to a project's palette, or recolor an existing one. Labels are lowercase. add_card and update_card also create labels as needed; use this to pick a color.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
name required	string	Label name, like "frontend". (max 24 chars)
color	string	Optional color. (one of slate, blue, violet, pink, orange, amber, green, teal)

update_label

Edit a label · needs a write token

Rename or recolor a label. A rename updates every card that has it; renaming onto another existing label merges the two.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
label required	string	The label to change. (max 24 chars)
name	string	New name. (max 24 chars)
color	string	Optional color. (one of slate, blue, violet, pink, orange, amber, green, teal)

remove_label

Remove a label · needs a write token

Delete a label from the project and take it off every card that has it.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
label required	string	The label to delete. (max 24 chars)

archive_card

Archive a card · needs a write token

Put a card away off the board without deleting it: finished work the human wants out of sight, or work set aside for later. It keeps everything, becomes read-only, and unarchive_card brings it back. Refuses a card another agent holds.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)

unarchive_card

Unarchive a card · needs a write token

Bring an archived card back to its lane on the board. find_cards with archived lists the archive.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)

delete_card

Delete a card · needs a token with Delete and restore cards

Move a card to the trash: duplicates, or work the human decided against. It can be restored with restore_card until the trash is emptied. Refuses a card another agent holds.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)

restore_card

Restore a card · needs a token with Delete and restore cards

Bring a deleted card back from the trash to its lane. find_cards with in_trash lists the trash.

Argument	Type	Description
card required	string	The card reference shown on the board, like "YK-12". (max 80 chars)

get_flow_rules

Read flow rules · needs a token with Manage lanes, swimlanes and flows

Read each swimlane's flow rules: what the board does when an agent claims, reports, asks, completes or releases a card (move it to a lane, flag it for the human), and auto-start rules (on entered: start an agent on a runner when a card enters `lane`; set up in the app). Rules run in order. Their ids are what update_flow_rule and remove_flow_rule take.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane	string	Swimlane name, like "Bugs". Optional; leave it out to look in every swimlane. (max 40 chars)

add_swimlane

Add a swimlane · needs a token with Manage lanes, swimlanes and flows

Add a swimlane (a row of the board for one kind of work) with its own lanes and flow rules, from a preset or as a copy of another swimlane's lanes and rules (without its cards).

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
name required	string	Name, like "Bugs". (max 40 chars)
preset	string	Starting lanes and rules: agentFlow (Ideas → Ready → Working → Needs you → Review → Shipped), bugTriage, research or simpleFlow (default). (one of agentFlow, bugTriage, research, simpleFlow)
copy_of	string	Copy this swimlane's lanes and rules instead. (max 40 chars)
position	integer	Optional 1-based position among the swimlanes (1 is first). Omit to leave it where it is. (1–50)
color	string	Optional color. (one of slate, blue, violet, pink, orange, amber, green, teal)

update_swimlane

Edit a swimlane · needs a token with Manage lanes, swimlanes and flows

Rename, describe, recolor or reorder a swimlane, turn epics on or off for its cards, or turn its Backlog on or off (turning it off moves parked cards into the first lane).

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane required	string	The swimlane to change. (max 40 chars)
name	string	New name. (max 40 chars)
description	string	What goes in it. Pass "" to clear. (max 280 chars)
epics	boolean	Whether its cards show and take epics.
backlog	boolean	Whether it has a Backlog for parked work.
position	integer	Optional 1-based position among the swimlanes (1 is first). Omit to leave it where it is. (1–50)
color	string	Optional color. (one of slate, blue, violet, pink, orange, amber, green, teal)

remove_swimlane

Remove a swimlane · needs a token with Manage lanes, swimlanes and flows

Delete a swimlane with its lanes and rules. Its cards (backlog included) move to a lane in another swimlane first, so nothing is lost. A board keeps at least one swimlane.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane required	string	The swimlane to delete. (max 40 chars)
move_cards_to_swimlane required	string	The swimlane its cards move to. (max 40 chars)
move_cards_to_lane	string	The lane there. Defaults to its first lane. (max 40 chars)

add_lane

Add a lane · needs a token with Manage lanes, swimlanes and flows

Add a lane (a column) to a swimlane's flow, at the end unless you give a position.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane	string	Swimlane name, like "Bugs". Optional; defaults to the first swimlane. (max 40 chars)
name required	string	Name, like "Review". (max 40 chars)
description	string	Optional: what the lane means. (max 280 chars)
wip_limit	integer	Optional work-in-progress limit, 1–99. (1–99)
done	boolean	Whether the lane holds finished work (Shipped, Fixed, Done): its cards stop counting as open.
position	integer	Optional 1-based position among the swimlane's lanes (1 is first). Omit to leave it where it is. (1–50)
color	string	Optional color. (one of slate, blue, violet, pink, orange, amber, green, teal)

update_lane

Edit a lane · needs a token with Manage lanes, swimlanes and flows

Rename, describe, recolor or reorder a lane, set its WIP limit (0 clears it), or mark it as a done lane.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane	string	The lane's swimlane, when the lane name isn't unique. (max 40 chars)
lane required	string	The lane to change. (max 40 chars)
name	string	New name. (max 40 chars)
description	string	What the lane means. Pass "" to clear. (max 280 chars)
wip_limit	integer	Work-in-progress limit, 1–99; 0 clears it. (0–99)
done	boolean	Whether the lane holds finished work (Shipped, Fixed, Done): its cards stop counting as open.
position	integer	Optional 1-based position among the swimlane's lanes (1 is first). Omit to leave it where it is. (1–50)
color	string	Optional color. (one of slate, blue, violet, pink, orange, amber, green, teal)

remove_lane

Remove a lane · needs a token with Manage lanes, swimlanes and flows

Delete a lane. Its cards move to another lane first (any swimlane), and rules that moved cards into it are switched off. A swimlane keeps at least one lane.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane	string	The lane's swimlane, when the lane name isn't unique. (max 40 chars)
lane required	string	The lane to delete. (max 40 chars)
move_cards_to required	string	The lane its cards move to. (max 40 chars)
move_cards_to_swimlane	string	That lane's swimlane. Defaults to the same swimlane. (max 40 chars)

add_flow_rule

Add a flow rule · needs a token with Manage lanes, swimlanes and flows

Add a rule to a swimlane: when an event happens to one of its cards (claimed, progress, needs_input, completed, released, stale), move the card to a lane of the same swimlane and/or flag it (owner: the human's turn; problem; clear). Rules run in order after the existing ones.

Argument	Type	Description
project	string	Project slug, name or card prefix. Optional; defaults to the project this session last worked in (claimed or filed a card in, or named to introduce). Without one and with several active projects, name the one you're working in: reads default to the first, but changes and get_next_card refuse to guess. An archived project is read-only: name it to read it. (max 80 chars)
swimlane	string	Swimlane name, like "Bugs". Optional; defaults to the first swimlane. (max 40 chars)
on required	string	The event that triggers it. (one of claimed, progress, needs_input, answered, completed, released, stale, pr_opened, pr_merged, pr_closed)
move_to	string	Optional lane in the same swimlane to move the card to. (max 40 chars)
set_attention	string	Optional flag to set: owner, problem, or clear. (one of owner, problem, clear)
enabled	boolean	Whether it runs (default true).

update_flow_rule

Edit a flow rule · needs a token with Manage lanes, swimlanes and flows

Change a flow rule by its id (from get_flow_rules): its event, target lane ("" to stop moving), flag ("" to stop flagging), order, or switch it on or off.

Argument	Type	Description
rule required	string	The rule's id. (max 80 chars)
on	string	The event that triggers it. (one of claimed, progress, needs_input, answered, completed, released, stale, pr_opened, pr_merged, pr_closed)
move_to	string	Lane in the same swimlane, or "" for none. (max 40 chars)
set_attention	string	owner, problem, clear, or "" for none. (one of owner, problem, clear, "")
enabled	boolean	Switch it on or off.
position	integer	Optional 1-based position among the swimlane's rules (1 is first). Omit to leave it where it is. (1–50)

remove_flow_rule

Remove a flow rule · needs a token with Manage lanes, swimlanes and flows

Delete a flow rule by its id (from get_flow_rules). Cards stay where they are; the event just stops moving or flagging them. To pause a rule instead, update_flow_rule with enabled: false.

Argument	Type	Description
rule required	string	The rule's id. (max 80 chars)