Klarum

Agent (AI Assistant, "Klara")

Generally available
Supplier and issuer
From basic

Klarum's conversational assistant: a tool-using ReAct agent on LangGraph that reasons over the org's own data, calls tools to read and act on it, searches the web, and drives the UI of the page the user is on.

Catalogue id
agent

Description

What it is

Klarum's conversational AI assistant, branded "Klara" to the end user. A tool-using ReAct agent built on LangGraph (create_react_agent) that holds a conversation, reasons over the firm's own data, calls tools to read and act on it, searches the web, and drives the UI of the page the user is on. Two modes: Chat (streaming multi-turn with tool use, citations, thinking/tool-call blocks) and Notebook (a Jupyter-style cell stack of promptable automations wired to the process engine + knowledge API). Context-scoped: firm-wide on its own page, but hard-pinned to a workspace + its issuer/tender when opened from that workspace.

What it helps users do

In natural language: find and triage the pipeline; discover and search live tenders; interrogate firm knowledge with citations and page numbers; read deep project context (details, company research, validation reports, history, dashboard analytics); take actions (update/delete projects, export to Excel, generate outreach emails, edit/save document content, write issuer-timeline notes (gated), convert transcripts into checklists); operate the current screen by text ("apply my template Noah", "reorder by deadline"); and run repeatable automations in Notebook mode. The pitch: instead of clicking dozens of screens, the consultant talks to one assistant that already knows their firm, workspace, the live market, and their documents, and can both answer and do.

Capabilities

  • Chat mode

    Generally available

    Streaming multi-turn conversation with tool use, citations and live thinking and tool-call blocks, persisted to process_runs so a session is resumable.

  • Notebook mode

    Generally available

    A Jupyter-style cell stack of promptable automations wired to the process engine and the knowledge API, for work that repeats.

  • Page-action bridge

    Generally available

    A server-stateless tool: the agent emits an action, the backend echoes it, and the frontend invokes the matching handler from its per-page registry. This is what lets the assistant operate the UI.

  • Contract manifest and state-edges

    Generally available

    A source-walked manifest maps every route to its components, testids, actions and dual doors, so an in-page view with no route of its own is still reachable and nothing is hand-maintained.

  • Cross-context isolation

    Generally available

    Row-level firm scoping on every SQL and RAG call plus a non-negotiable prompt rule that the agent may only answer about the pinned workspace, so one deal cannot leak into another.

  • Gated write tools

    Generally available

    Write tools sit behind a per-org opt-in flag and refuse when mis-wired, and confirm-gated proposals put a human between an intention and an effect.

  • Brand-aware document generation

    Generally available

    On request the agent produces a finished on-brand PDF dressed in the org's own design tokens, routing between approved brands by matching the new document to the org's past accepted ones.

Integrations

  • Tavily Web search and company research behind the WebSearch tool and a heavier research agent, registered only when an API key is set.

Keywords

langgraph, react-agent, tools, sse, notebook, page-bridge, dspy, rag, citations

Provenance

This page renders a published artifact. Read the same records as JSON if you would rather check the source than the page.