Summary: Intuition is best understood not as a single app, but as a token-curated semantic graph that turns entities, claims, and confidence into reusable onchain objects. Its core mechanism is the separation between identifiers (Atoms), relationships (Triples), and economically weighted conviction (Signals), plus an incentive system meant to push users toward canonical identifiers and shared schemas instead of fragmented app-specific data silos. That makes it a useful comparison class for Ethereum Attestation Service, Sign Protocol, OpenRank, and portable social-graph systems: those systems standardize attestations or compute over trust data, while Intuition tries to make the underlying knowledge graph itself economically legible and ownable.
What it does:
Represents any person, object, concept, document, or byte string as an Atom, a persistent onchain identifier
Represents relationships and claims as Triples in subject-predicate-object form, where each component is itself an Atom
Lets users add Signals by staking behind Atoms or Triples to express confidence, support, or counter-position
Uses bonding curves and token incentives to encourage convergence on shared canonical identifiers and schemas rather than duplicate fragmented entries
Exposes SDK, GraphQL, documentation, and network infrastructure for applications, agents, and data systems to read from and build on the graph
Positions the resulting graph as a portable trust, identity, attestation, and knowledge layer that can be reused across apps, chains, and AI systems
Key claims:
The official docs repeatedly frame Intuition as a protocol for decentralizing information rather than only identity or reputation. The analytically important move is to treat knowledge objects themselves as protocol-native assets.
The Atoms / Triples / Signals split is the main reusable mechanism. Intuition separates naming, relationship expression, and conviction-weighting instead of collapsing them into one attestation primitive.
The primitives docs make explicit that Triples can reference other Triples as Atoms, which means the protocol is trying to support recursive, composable claim structures rather than a flat credential ledger.
The economics docs are especially important because they show Intuition is not merely a semantic-data standard. Bonding curves, early-mover rewards, and standards-aligned incentives are meant to govern which identifiers and predicates become canonical.
This makes Intuition a strong comparison class for EAS and Sign Protocol. Those systems focus on issuing attestations, while Intuition is trying to create a market structure around identifier convergence, claim reuse, and graph legibility.
Intuition also overlaps with OpenRank and Human Passport from the other side of the stack: instead of computing scores over an existing trust graph, it tries to make the graph and its weighting substrate economically native from the start.
The project belongs in the corpus because it surfaces a distinct control question: if many parties can name the same thing and express related claims, what mechanisms decide which identifiers, predicates, and graph fragments become canonical enough for everyone else to build on?
Whitepaper: Intuition publishes an official whitepaper and whitepapers portal. The strongest primary materials reviewed here were the official site, intro docs, primitives docs, economics docs, documentation repository, and the official whitepaper PDF; see ../whitepapers/intuition-primary-sources-2026-05-10.md and ../whitepapers/intuition-whitepaper.pdf.