Editing a flow needs a larger screen. This is the read-only view. Back to research.

Literature landscape

Cluster the gathered evidence into a map and a timeline, then read the landscape - where the field is dense and where it is thin.

v1published

Orchestrator

Orchestrator

the one agent the researcher talks to · owns add_node / draw_edge

balancedbudget 10

You are the researcher's co-pilot. You propose; the researcher decides. Delegate to your sub-agents; every graph mutation is a proposal the researcher commits.

Sub-agents · called as tools

Scout

scoutfast

Gather broadly across the measured corpus and the linked literature to feed the map.

derive_query_plansearch
auto_synonym_expansionon

Cartographer

cartographerbalanced

Cluster the gathered evidence into a literature map; lay it out on a timeline so the field's shape reads at a glance.

build_map
timeline_map_defaulton

Synthesizer

synthesizerdeep

Read the landscape: the dense clusters, the thin spots, and the gaps. Keep and flag ungrounded points.

synthesize

Output template · the deep-research deliverable

  1. 1Landscape overview

    The major evidence clusters and how they relate.

  2. 2Timeline

    How the field's activity is distributed over time.

  3. 3Gaps

    Where the landscape is thin - the openings for new work.

Trust floor

non-negotiable

Locked. No template or workspace can weaken these - they are code, not configuration.

  • Grounding - the deterministic firewall checks every cited claim
  • Attribution - every node is owned (prov:wasAttributedTo)
  • Human gates - approve the plan, commit each node and edge
  • Budget - the Orchestrator LLM-call ceiling
approve_query_plancommit_nodecommit_edgemaxLlmCalls≤10