Skip to main content
Read the workspace’s shared work record: every pitch, listing, reply, published page, fix, note, and to-do with its status, target, owner, due date, latest note, and what changed after (outcomes show correlation, not proof of causation). Filter by brand, status, kinds, target, contact, topic, owner, goal, due, or since; pass item for one item’s full trail, or view: “to_file” / “contacts”. Free.

Who may call it, and what it costs

Price. Free. Every reply says what the call cost in meta.costCredits. The prices are on Spending, tracking and stopping.

Arguments

string
brd_ id
list of one of planned, drafted, sent, no_reply, replied, won, declined, published, fixed, live, shipped, noted, done, suggested, open, reported, ended
open (not ended), reported (an agent said won or live, monitoring has not agreed), ended, suggested (the tasks Parse suggests), or any status up to 17
boolean
with suggested: only the tasks another assistant holds
list of one of pitch, listing_request, reply, page_written, page_edit, correction, site_fix, note, pitched, pitch_sent, listed, declined, published, fixed
item kinds; the older log_work names (pitched, published, listed, declined, fixed) read as their kinds up to 15
string
a page (URL or pg_ id), a site (domain or site:), a contact (ct_ id or email) or a top_ id: items aimed at it
string
the same as target, as older calls pass it
string
ct_ id or email: items to or about that contact
string
top_ id, or words of the gap
string
string
gol_ id: the work filed under it
one of overdue, today, week or date
open items due by then: overdue, today, week (the next 7 days) or a date
date
YYYY-MM-DD: items changed on or after it
string
itm_ id: that item with its full trail and notes
one of items, to_file, contacts
items (default); to_file: notes Parse could not anchor; contacts: the workspace’s contacts
integer
default 25; the token cap wins min 1, max 200
string
Every argument the tool takes, as GET /v1/tools serves it, every shared object written out; https://agent.parse.gl/openapi.json serves the same schema with the filter and the window named once under components and pointed to by $ref. MCP tools/list leaves out the arguments few calls send, and says the filter and the window once, under search_answers; resolve_ids {"kinds": ["argument"]} lists what it left out, and a call with any of them runs:

Example call

The open work, with what falls due in the next 7 days. Over REST:
Over MCP, the assistant calls the tool by name with the same arguments:
More calls:
  • Everything done on a page, by every brand of the workspace, with what happened after: {"target":"https://example.com/best-tools"}
  • One item’s trail and notes: {"item":"itm_…"}
  • Notes Parse kept to file, each with the one thing it needs: {"view":"to_file"}
The reply has data, context, next, page and meta; follow next for the calls that come after. A refused call returns { code, message, fix }: see the error table.