Skip to main content
GET
Read what a conversation may apply without asking

Authorizations

X-API-Key
string
header
required

Path Parameters

conversation_id
string<uuid>
required

Response

Successful Response

What one conversation may apply without asking: the list, and where it comes from.

Read for the shield in the chat header. follows_defaults tells the header whether to show the list as the workspace's (change it in Settings) or as this thread's own (take one back here, or return to the defaults). granted_tools is what the shield lists: the AI Assistant's cards are decided by tool, and the verbs decide only an agent's.

follows_defaults
boolean
required

Whether the list is the workspace defaults, read live. False once the conversation has a list of its own, which a change in Settings no longer reaches.

granted_tools
string[]
required

The write operations the AI Assistant runs without asking in this conversation, in effect now: operation names as GET /assistant/tools/ lists them, and family:<area> for the rest of one part of the product. Approving a card for this conversation adds its operation here. Only what still allows something is listed, by the name its operation has today.

granted_verbs
enum<string>[]
required

The verbs this conversation applies without asking, in effect now.

The eight things an AI may do to a workspace.

Exactly eight, not sixty: today's cell-level actions all run against one cell, and building a table, adding contacts to a sequence and starting a search had no home at all. Anything new here is a product decision about what an assistant may touch, not a convenience for a planner.

Available options:
create_table_from_search,
add_column,
run_column,
enrich_rows,
add_contacts_to_sequence,
push_to_destination,
call_endpoint,
start_search