> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sleuthintel.io/llms.txt
> Use this file to discover all available pages before exploring further.
# Tree of Knowledge: Sleuth Intel's Intelligence Graph
> Learn how Sleuth Intel's verified intelligence graph is structured across 10 niches, verified through machine and judgment lanes, and consumed by users.
The Tree of Knowledge is Sleuth Intel's verified intelligence graph — the structured, curated layer of on-chain and off-chain intelligence that platform users actually consume. It maps wallets, accounts, traders, and narratives as nodes, and the relationships between them as edges. The Tree does not accept raw submissions directly; every piece of intelligence passes through a separate intake surface before it can be promoted to the Tree.
## The Belt vs The Tree
**Intake surface — pre-acceptance**
Raw submissions land here first. Content is public and unverified. The Belt is intentionally messy — it is the working layer where contributors submit, verifiers evaluate, and machine checks run. Nothing on The Belt should be treated as confirmed intelligence.
**Verified intelligence — what users consume**
Submissions that pass verification are promoted to The Tree. This is the curated, trustworthy layer. Every node and edge here has passed either a machine verification check or a human judgment panel. Users and downstream integrations read from The Tree.
Think of The Belt as the factory floor and The Tree as the finished product shelf. Submissions move from one to the other only after passing verification.
## The 10 Intelligence Niches
Every submission targets exactly one niche. Niches define what kind of intelligence a node represents and which verification standards apply.
| Niche | What It Covers |
| ----------------------- | -------------------------------------------------------------------- |
| **Wallet Identity** | Who controls a wallet — names, entities, attribution |
| **Wallet Behavior** | How a wallet trades — patterns, strategies, timing tendencies |
| **Contract / Protocol** | Smart contract metadata, protocol classification, risk flags |
| **Risk / Threat** | Scams, exploits, rug pulls, malicious actors |
| **PM Trader** | Prediction market participant profiles and track records |
| **PM Market** | Prediction market metadata, resolution conditions, category tags |
| **Social Account** | CT accounts, Telegram, Discord — identity and affiliation links |
| **Account Accuracy** | Verified track record of social accounts making on-chain predictions |
| **Narrative / Theme** | Macro and micro narratives shaping market behaviour |
| **Project** | Token projects, DAOs, protocols — entity-level intelligence |
## Cross-Domain Links: The Most Valuable Intel
Linking a node in one niche to a node in another niche is called a cross-domain link. These are rarer and harder to establish — but they are the most valuable contributions in the Tree.
**Example:** Connecting a wallet address (Wallet Identity) to a Crypto Twitter account (Social Account) to a prediction market trader profile (PM Trader) creates a three-node chain that identifies a real person's activity across three different surfaces.
Cross-domain links are the rarest submissions on the platform and receive the highest grade weightings. If you can establish a credible wallet ↔ CT account ↔ PM trader connection with supporting evidence, prioritise it.
Why are they valuable? Because most intelligence lives in silos. Wallets don't self-identify. Social accounts don't post their wallet addresses. Prediction market profiles don't link to Twitter. The intelligence gap between siloed data and connected data is enormous — and the Tree closes it.
## Two Verification Lanes
Not all submissions travel the same path from The Belt to The Tree. The lane is determined by whether the claim is programmatically verifiable.
**Instant, on-chain verifiable claims**
If a claim can be confirmed by querying on-chain state — transaction history, contract bytecode, token balances — it routes to the machine lane. Verification is near-instant and fully automated. No human panel is assembled.
**Contextual claims requiring human evaluation**
Claims that cannot be confirmed programmatically — social identity links, behavioural attribution, narrative classification — route to the judgment lane. Three verifiers are selected at random, cast blind verdicts within a 5-day window, and the outcome is determined by consensus.
Blind verdicts mean verifiers cannot see each other's answers while voting. This prevents herding and anchoring bias, keeping each verdict independent.
## Anti-Spam: The \$10 Activation Fee
Before you can submit your first piece of intelligence, you pay a **one-time \$10 SLEUTH activation fee**. This is a permanent unlock — you pay it once, and it covers all future submissions.
The activation fee is not a recurring cost. It is a one-time friction mechanism designed to prevent throwaway account submissions. Once activated, your account remains active permanently.
Beyond the activation fee, the platform uses additional anti-spam controls:
* **Submission slots** — you hold between 3 and 25 open slots at any time, determined by your track record
* **Rejection streak cooldowns** — consecutive rejections trigger escalating cooldown periods before you can submit again
Both mechanisms mean that contributor quality directly determines contributor capacity. Build a strong track record and your submission throughput increases automatically.