> For the complete documentation index, see [llms.txt](https://solyx.gitbook.io/solyx-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://solyx.gitbook.io/solyx-docs/system-design.md). # System Design ### 3.1 Trust Scoring Engine Overview Solyx evaluates wallets by observing their on-chain activity over time. Each wallet receives a trust score ranging from 0 to 100, as well as optional behavioral tags that describe observed patterns. ### Inputs * Transaction history: trades, swaps, mints, voting * Smart contract interactions: across DeFi, DAOs, and beyond * Time-based behavior: frequency, burstiness, consistency, session velocity * Graph signals: wallet connections, shared fund flows, sybil clusters * Airdrop behavior: farming, sniping, ghosting vs. engagement ### Outputs * **Trust Score (0–100)**: Continuously updated based on behavior * **Behavioral Tags**: Profile descriptors like "Sniper Bot," "Whale," "sybil-like" * **Annotations**: Optional user-submitted flags or insights attached to any wallet These outputs allow protocols to make better decisions around access, weight, filtering, and community alignment. *** ## 3.2 Wallet Identity Model Solyx does not rely on static identifiers or user-submitted credentials. Instead, it infers identity through the behavioral patterns and relational structures present in wallet activity. The wallet identity model is built around the idea of "behavioral identity", a unique, probabilistic signature that emerges from consistent actions over time. Each wallet is evaluated based on its transactional fingerprint, including the types of interactions it engages in, its contract activity, timing patterns, and velocity of movement. This data is used to form clusters of wallets that are likely operated by the same entity or follow similar behavioral logic. For example, wallets that interact with each other frequently, follow the same transaction timing, or operate in a highly synchronized manner may be treated as a sybil cluster or as part of an organizational unit. Relationship mapping also helps surface larger behavioral structures in the ecosystem. If Wallet A is consistently funded by Wallet B, which is also linked to Wallets C and D through shared interactions, Solyx will surface this relational web as part of its identity inference. This allows for richer coordination filters and sybil resistance without requiring deanonymization or KYC. *** ## 3.3 Behavioral Tagging System Behavioral tagging in Solyx serves as a semantic layer on top of raw scoring. While the trust score is a numerical representation of behavioral credibility, tags offer interpretable labels that describe the likely role or function of a wallet within the ecosystem. Tags are assigned automatically based on pattern detection across time. For example, a wallet that repeatedly instantly snipes tokens as they are created and exits quickly may receive a tag like "Sniper Bot." A wallet that moves in sync with privileged wallets might be flagged as an "Insider." Wallets that consistently hold tokens over a long period of time through volatility could be tagged as "Long-Term Holder." Accounts that recycle the same farming patterns might be identified as "Copytrader Farmer." These tags are dynamic. They evolve as behavior shifts and are recalculated periodically to reflect the latest context. Some tags may also carry additional scoring weight. For instance, a tag like "Sybil Cluster" could lower a wallet’s reputation, while "Consistent Holder" could enhance it. However, tag effects are modular and tunable depending on the protocol or application consuming the data. By offering a categorical view of wallet behavior, tags enable more nuanced filtering and access decisions, helping protocols distinguish extractive activity from valuable contributions. *** ## 3.4 Annotation Layer In addition to algorithmic trust scoring and behavioral tagging, Solyx introduces a human-readable layer through wallet annotations. These are public notes by users that any user can leave on a wallet to provide context, caution, or insight. Annotations allow the community to participate in surfacing useful information that may not be captured by models alone. For instance, a user may flag a wallet as a known exploiter, vouch for someone they’ve transacted with in a DAO, or document wallet behavior that’s emerging but not yet statistically common. To preserve credibility and prevent abuse, annotations require signature verification, and the system supports reporting, counter-annotations, and upvote/downvote moderation. While annotations do not directly affect trust scores, they serve as powerful contextual signals, giving protocols and other users another layer of signal to base coordination on. In the long term, the annotation layer opens the door for reputation graphs and decentralized review networks, helping Solyx evolve from passive scoring to participatory intelligence. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://solyx.gitbook.io/solyx-docs/system-design.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.