OloLand DD — Due Diligence Plugin for Claude Code

The first punch + second punch for M&A due diligence.

OloLand structures the chaos of a data room into deterministic financial models, risk taxonomies, and cross-deal intelligence. Claude reasons over the result. Together they produce institutional-grade due diligence that neither can achieve alone.

Install

claude plugin add github:ololand-ai/ololand-dd-plugin

Quick Start

Install the plugin (see above).
Connect your account — open ololand.ai/connect, sign in, and copy your agent key.

Set the key in your environment:

export OLOLAND_AGENT_KEY=olo_agent_sk_...

Run your first analysis:
```
/dd-analyze
```

Commands

Command	Description
`/dd-analyze`	Run full due diligence on a deal — risk, valuation, forensic QoE, and investment memo.
`/risk-report`	Generate a structured risk report across OloLand's 246-category taxonomy.
`/valuation`	Run DCF, LBO, comparable transactions, and Monte Carlo valuation models.
`/similar-deals`	Find historically similar deals and surface cross-deal patterns.
`/deal-search`	Search across all documents in a deal's data room with hybrid vector + keyword search.
`/war-game`	Simulate competitive strategy scenarios using RL-powered market dynamics.
`/talk-to-deal`	Ask a natural-language question about any deal and get a sourced answer.

What's Different from Raw Claude

Using Claude alone on a data room is like reading every document yourself — thorough but slow, with no structure and no memory across deals. OloLand + Claude is a one-two punch:

First Punch (OloLand)

Before Claude sees a single token, OloLand has already:

Classified risk across a 246-category taxonomy with a fine-tuned Qwen 3 4B model
Indexed every document into a hybrid vector + sparse search index with cross-document reconciliation
Built a knowledge graph linking entities, claims, and financial figures across hundreds of files
Run forensic QoE — Beneish M-Score, Benford's Law, and revenue/expense reconciliation to flag manipulation
Computed valuations — DCF, LBO, Monte Carlo, comparable transactions, and real options with deterministic engines (not LLM math)
Matched against prior deals — cross-deal learning surfaces patterns from every deal your firm has analyzed

Second Punch (Claude)

Claude then reasons over a filtered, structured subset — not raw PDFs. It synthesizes findings, identifies what matters, and produces investment memos grounded in OloLand's deterministic outputs.

The Flywheel

Every deal makes the system smarter. Analyst corrections refine risk models. Outcome tracking calibrates predictions. Cross-deal patterns compound. This is institutional memory that no single-session LLM can replicate.

Benchmark

OloLand scores 90.5% vs Claude's 88.5% on the Gauntlet v4 T5 institutional due diligence evaluation (dual-judge scoring by Gemini 3.1 Pro + Claude Opus 4.6). The gap widens on forensic, reconciliation, and visual-decision tasks.

MCP Tools (33)

The plugin connects to OloLand's MCP server, which exposes 33 tools grouped by category:

Category	Tools
Deal Intelligence	`list_deals`, `get_deal`, `get_deal_summary_tiles`, `get_deal_indicators`
Financial Valuation	`get_financial_snapshot`, `get_dcf_valuation`, `run_monte_carlo_simulation`
Risk Analysis	`get_deal_risks`, `get_evidence_links`
Documents	`list_deal_documents`, `search_deal_documents`
Knowledge Graph	`query_knowledge_graph`, `get_entity_neighbors`, `search_knowledge_graph`
Cross-Deal Learning	`find_similar_deals`
Reports	`generate_investment_memo`, `generate_cim`, `export_deal_dossier`
Market Intelligence	`research_market`, `deep_market_research`, `search_pe_buyers`, `search_targets`, `search_ma_deals`
Strategy	`run_war_game_simulation`, `analyze_build_vs_buy`, `generate_acquisition_thesis`
Corp Dev	`batch_triage_companies`
Voice	`talk_to_deal`
Workflow	`run_due_diligence`, `check_task_status`, `decompose_intent`, `get_plan_status`, `list_missions`

ololand-dd

Quick Install