Skip to main content
Back to challenges
Challenge

Build a Fund Prospectus Compliance Audit Engine with LangChain and Windsurf

Develop a compliance verification agent using LangChain and Windsurf IDE workflows to audit asset management fund prospectuses against MAS regulatory directives. Extract investment mandate rules and detect clause contradictions in real time.

Workflow AutomationHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Utilize LangChain LangGraph agent workflows developed with Windsurf developer assistance to parse, verify, and highlight non-compliant fund prospectus terms.

Delivery guide

How work is evaluated

Evaluation

Evaluates detection accuracy of non-compliant clauses in simulated fund prospectus texts.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

violation_detected

Verifies that excessive concentration is flagged as non_compliant

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

citation_precision

Precision of extracted text citations • target: 0.95 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Construct stateful compliance checking graphs using LangChain's LangGraph framework

  • Optimize local prompt engineering and workflow iterations using Windsurf AI engineering tools

  • Extract structured investment limit rules from unstructured regulatory PDF texts

  • Generate automated compliance violation matrix reports with precise citations

Resources and assets

Reference links and supporting material

Dataset notes

Dataset containing 20 fund prospectus excerpts with synthetic regulatory violations.

Challenge setup

Evaluates detection accuracy of non-compliant clauses in simulated fund prospectus texts.

Challenge input

JSON object with prospectus_text and rule_set_version

LangChain

Provides LangGraph for building stateful compliance workflows.

Windsurf

Accelerates agent development and prompt iteration.

Evaluated output

JSON object with compliance_status, violations array, and source_clause_citations

Evaluation checks
  • Verifies that excessive concentration is flagged as non_compliant
  • Precision of extracted text citations • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: LegalBench Regulatory Compliance
  • Citation Precision target: 0.95
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • LangChain · Required
  • Langchain · Optional
  • Windsurf · Optional
Observation
  • Windsurf · Optional
Orchestration
  • LangChain · Required
  • Langchain · Optional

Evaluation contract

  • violation_detected · Weight 1
  • citation_precision · Weight 1

Metrics

  • citation_precision · Target 0.95

Benchmarks

  • LegalBench Regulatory Compliance
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with prospectus_text and rule_set_version

    Output: JSON object with compliance_status, violations array, and source_clause_citations

    Execution harness: Python

Evaluation tools

  • LangChain · Agent Chains & Graphs
  • Windsurf · AI Engineering Environment

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Fund Prospectus Compliance Audit Engine with LangChain and Windsurf