Catalyst evidence graph

Reference-database relations (hypotheses)

get_relations

Read-onlyIdempotent

Lists what public reference databases (Reactome, UniProt, Rhea, ChEBI and others) record about one node by id, from search_nodes: transporters, enzymes, pathways, expression. Every row is assertion_class 'inferred': a hypothesis about how a compound could act, not a result anyone measured in people. Present these as possible mechanisms and never as effects. For effects, use find_evidence. For the reactions a compound or enzyme takes part in, use get_reactions; for what a compound can become, use reach. predicate limits the list to one relation type, named as in the predicate field of a row (for example transported_by for transporters, metabolized_by for enzymes). Returns total and up to limit rows (default 10); offset pages the rest, and the default is enough to answer. A page too large for one result is cut to fit, and truncated with next_offset says how to get the rest. An empty list with a predicate means the databases record no relation of that type for this node. That is a complete answer, so do not repeat the call with the same predicate. Leave predicate out once only when the person asked broadly and you need to see which types the node has.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "node_id": "<node_id>"
    },
    "name": "get_relations"
  }
}

Input Schema

Name	Required	Description
`limit`	No	Rows to return (1 to 50).
`offset`	No	Rows to skip, to page past limit.
`node_id`	Yes	A node id in the form PREFIX:local, e.g. CHEBI:16919 (creatine). Get it from the id field of a search_nodes hit.
`predicate`	No	Limit to one relation type, as in the predicate field of a row, e.g. transported_by.

Schema Changelog

Changes observed during successful MCP inspections.

Changed8 schema fields changedNov 5, 2026

changedInput schema / properties / limit / default
```
Previous value: -25New value: +10
```
addedInput schema / properties / limit / description
```
Added value: +"Rows to return (1 to 50)."
```

changedInput schema / properties / node_id / description

Previous value: -"A node id from search_nodes, e.g. CHEBI:16919 (creatine)."New value: +"A node id in the form PREFIX:local, e.g. CHEBI:16919 (creatine). Get it from the id field of a search_nodes hit."

addedInput schema / properties / offset / description
```
Added value: +"Rows to skip, to page past limit."
```

changedInput schema / properties / predicate / description

Previous value: -"Limit to one relation type, e.g. transports."New value: +"Limit to one relation type, as in the predicate field of a row, e.g. transported_by."

addedInput schema / properties / predicate / enum

Added value: +[
+  "transported_by",
+  "substrate_of",
+  "inhibits",
+  "activates",
+  "binds",
+  "metabolized_by",
+  "converted_to",
+  "participates_in",
+  "regulates",
+  "expressed_in",
+  "absorbed_at",
+  "accumulates_in",
+  "affects",
+  "depletes",
+  "requires_cofactor",
+  "contraindicated_with",
+  "alters_exposure_of",
+  "contradicts",
+  "has_adi",
+  "response_varies_by",
+  "has_functional_parent",
+  "enantiomer_of",
+  "has_hepatotoxicity_likelihood",
+  "regulates_expression_of",
+  "interacts_with",
+  "covaries_with",
+  "affects_phenotype",
+  "has_phenotype",
+  "has_germline_basis_in",
+  "located_in",
+  "part_of",
+  "affects_relative_to"
+]

removedInput schema / properties / predicate / maxLength
```
Removed value: -40
```
removedInput schema / properties / predicate / pattern
```
Removed value: -"^[a-z_]+$"
```

Changed1 schema field changedOct 4, 2026
- addedInput schema / additionalProperties
```
Added value: +{}
```
First observedOct 4, 2026

TDQS

A4.4/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so safety is covered. The description adds genuine context beyond that: rows are inferred hypotheses that must be framed as possible mechanisms, plus truncation/next_offset behavior and the epistemic meaning of an empty predicate result. It does not describe row shape or field contents, which is a remaining gap given no output schema.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core action, then warnings, then sibling routing, then parameter behavior; every sentence carries operational meaning. It is dense but long, with the predicate explanation spread across two separated passages, which slightly hurts scanability.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema, the description still covers the return envelope (total, up to limit rows, default 10, offset paging, inlined next_offset on truncation) and the interpretive rule for empty results. That is everything an agent needs to call it and read the response correctly without a second round trip.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so a baseline 3 is guaranteed; the description earns more by explaining how to pick a predicate value ('named as in the predicate field of a row', with transported_by and metabolized_by examples) and by stating the default limit is usually enough and offset pages the rest. It stops short of describing every enum member, which is fine because the enum itself is self-documenting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Names the specific verb and resource ('Lists what public reference databases ... record about one node by id') and scopes it to the DBs by name (Reactome, UniProt, Rhea, ChEBI). It also draws the decisive line that every row is assertion_class 'inferred', which separates this from the evidence-oriented siblings.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit routing: use find_evidence for effects, get_reactions for reactions, reach for what a compound can become, and get the id from search_nodes. It also states when to omit predicate (only for a broad question to see which types exist) and warns not to repeat a call after an empty predicate result.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Catalyst evidence graph tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.