Streaming 24h · Cash session shut · Connecting · --:--:-- UTC

USD/INR—0.00%EUR/INR—0.00%GBP/INR—0.00%JPY/INR—0.00%AUD/INR—0.00%SGD/INR—0.00%CHF/INR—0.00%HKD/INR—0.00%
Reliance Industries → Singapore · ₹475 cr · ODI

Agent graph

Many small agents. One reviewable decision.

Ask the rail a statutory question. A coordinator plans the run, a retriever pulls the provisions, a grader decides whether they are good enough, and only then does the filing agent draft. A separate quality agent checks every citation back to a retrieved passage before the answer reaches the desk — and every step is timed, counted and costed.

Multi-agent graph

No single prompt does everything. A coordinator routes work to a retriever, a grader, a rewriter, a filing drafter and a quality assurance agent, each with one job and one contract.

Corrective RAG

Semantic retrieval is graded before drafting. Weak evidence rewrites the search and expands the local corpus; no external statutory evidence is implied.

LCEL nodes

Every node is a composed prompt-to-model chain consumed as a stream, so long generations flow instead of stalling, and nodes run asynchronously inside the graph.

Observation layer

Each span carries latency, token counts and cost. Runs export to JSON and CSV, and mirror to LangSmith when a workspace key is present.

Graph intake

How the graph routes

The coordinator hands the question to retrieval; the CRAG grader decides the next edge. Partial evidence goes through the query rewriter and back to retrieval; still-poor evidence expands the local corpus. The drafter writes the filing packet and QA must pass it before the composer answers — malformed QA fails closed.

StateGraph — live topology & trace replay

no run yet
relevantpartialre-searchstill poorverdictGraph CoordinatoridleData RetrieveridleRetrieval Grader (CRAG)idleCoding / FilingidleQuery RewriteridleExpanded CorpusidleQuality AssuranceidleResponse Composeridle

Node inspector

Run the graph, then click any node to see what that agent did. Dashed edges are conditional routes chosen by the grader.

Managed deployment and graph testing

The graph runs through the application server. Managed deployment is configured as a future connection, not an active service. The suite scores application graph outputs on structured filing packets; exported results are evidence for review, not legal approval.

Managed deployment — LangGraph Platform

graph “statutory” · langgraph.json

Runtime

Application-hosted LangGraph via server functions. Managed LangGraph deployment is not connected.

Scaling

Request-driven application runtime; no managed graph queue or background resume is provisioned.

Persistence

Verified users: private append-only run outcomes in Lovable Cloud. Node-resume checkpoints are not configured.

Observability

Per-run spans in the desk and private run history; LangSmith mirroring requires a configured credential.

Assistant configurations · managed deployment not connected

  • Statutory determination

    asst_statutory_full

    Full run: coordinate, retrieve, grade, draft the filing packet, quality-check every citation.

    recursion_limit 24 · streaming on · corrective retries 2

  • Corrective retrieval probe

    asst_retrieval_only

    Retriever + grader + rewriter only — used to test the corpus without spending draft tokens.

    interrupt_after ['grader'] · streaming on

  • Filing QA gate

    asst_qa_gate

    Re-runs the quality agent against an existing draft before a filing leaves the desk.

    entry ['qa'] · deterministic temperature 0

The factory at src/lib/agents/platform-graph.server.ts:makeDeployedGraph is the same builder the desk runs in-process, so a managed run and a desk run execute an identical topology.

Graph regression suite — 4 cases against the application graph

Not run in this session

How an agent answer reaches a filing

Treasury desk — 3D process simulation

How a treasurer's action becomes a supervised, rate-stamped record.

Live USD/INR

₹86.9565 ▬

Tape status

Syncing

Print sequence

#0

Paper today → rail

24 days → 9 min

Who acts

Treasury

Rule 7 headroom

What happens at this stage

Existing commitments and headroom are read from the ledger, not rebuilt in a spreadsheet.

Elapsed 30s of 130s on the rail · stage cost 30s

Qwen · reasoning

Statutory reasoning model reads the rule and the filing together, and writes the conclusion with the rule quoted.

LangChain · orchestration

Chains the retrieval, tool calls and checks in a fixed order so every filing is examined the same way.

Alibaba Cloud · compute

Runs the heavy document and screening workloads in an India-resident region.

Federated ledger

Each result is hash-chained into a block every regulator node holds a copy of.

Human desk

A named officer signs this off. The machine only prepares the file.