Honcho
for LuumenAI
Read peer memory and session history, then write conclusions as approved steps
Honcho holds the memory behind your stateful agents: workspaces, peers, sessions, messages, and the conclusions drawn from them. Connect it and Luumen can read that history directly — list sessions, search messages, pull a peer's representation — then record new conclusions or messages as previewed steps you approve first. Authentication is by API key.
The Honcho toolbox
16 tools: 10 read, 6 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadChat With PeerRun usage-bearing agentic search and reasoning over a peer's learned representation to answer a question.
- ReadGet MessageReturn one message by its exact ID within a session.
- ReadGet Peer RepresentationReturn a curated subset of conclusions representing a peer.
- ReadList ConclusionsReturn one page of stored conclusions in a workspace; use Search Conclusions for semantic retrieval.
- ReadList MessagesReturn one page of messages in a session; use Get Message for exact-ID retrieval.
- ReadList PeersReturn one page of peers in a workspace.
- ReadList SessionsReturn one page of sessions in a workspace.
- ReadList WorkspacesReturn one page of Honcho workspaces visible to the connected admin key.
- ReadSearch ConclusionsSemantically search stored conclusions in a workspace.
- ReadSearch MessagesSemantically search messages at workspace, peer, or session scope.
- WriteCreate ConclusionsStore a batch of 1-100 explicit conclusions about peers, optionally tied to sessions. Approval by default
- WriteCreate MessagesAdd a batch of 1-100 attributed messages to a Honcho session. Approval by default
- WriteDelete SessionPermanently delete one known session and its session-scoped data. Approval by default
- WriteGet or Create PeerReturn a peer by ID within a workspace, creating it with the supplied metadata and configuration when absent. Approval by default
- WriteGet or Create SessionReturn a session by ID, or create it with initial peers, metadata, and configuration when absent. Approval by default
- WriteGet or Create WorkspaceReturn a workspace by ID, or create it with metadata and configuration when absent. Approval by default
One prompt, start to finish
What a governed Honcho run looks like inside Luumen.
Questions
Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.
Read actions answer immediately. Anything that writes — create conclusions, create messages, delete session, get or create peer, and more — is shown as a plan and requires approval by default, including the 1 action classified as destructive. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.
You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.
Every call to Honcho — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.
Often connected alongside
All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task status
AffindaRead parsed documents, collections, and annotations — change them with approval
Agent MailGive the agent its own mailbox for reading threads and sending updates
AI/ML APIInspect assistants, threads, and runs — and change them as approved stepsPut Honcho to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.