Quick lookup documentation in Redis-style format. Each entry includes syntax, description, examples, and notes.
| Document | Description |
|---|---|
| Commands | All meta commands (.kg, .rule, .rel, .load, etc.) |
| Functions | All 55 builtin functions |
| Syntax Cheatsheet | One-page syntax reference |
| Command | Description |
|---|---|
.kg |
Knowledge graph operations |
.kg list |
List all knowledge graphs |
.kg use <name> |
Switch to knowledge graph |
.kg create <name> |
Create knowledge graph |
.kg drop <name> |
Delete knowledge graph |
.rel |
List all relations |
.rel <name> |
Show relation schema |
.rule |
List all rules |
.rule def <name> |
Show rule definition |
.rule remove <name> <idx> |
Remove rule clause |
.rule drop <name> |
Remove entire rule |
.session |
Show session state |
.session clear |
Clear session |
.load <file> |
Load script file |
.status |
Show system status |
.compact |
Compact storage |
.help |
Show help |
.quit |
Exit |
| Aggregation | Syntax | Description |
|---|---|---|
| COUNT | count<X> |
Count distinct values |
| SUM | sum<X> |
Sum numeric values |
| MIN | min<X> |
Minimum value |
| MAX | max<X> |
Maximum value |
| AVG | avg<X> |
Average value |
| COUNT_DISTINCT | count_distinct<X> |
Count distinct values |
| TOP_K | top_k<K, ..., OrderVar:desc> |
Top K by ordering |
| Operator | Meaning |
|---|---|
= |
Equal |
!= |
Not equal |
< |
Less than |
<= |
Less than or equal |
> |
Greater than |
>= |
Greater than or equal |
| Operator | Meaning |
|---|---|
+ |
Addition |
- |
Subtraction |
* |
Multiplication |
/ |
Division |
% |
Modulo |
| Category | Count | Examples |
|---|---|---|
| Distance | 4 | euclidean, cosine, dot, manhattan |
| Vector Ops | 4 | normalize, vec_dim, vec_add, vec_scale |
| LSH | 3 | lsh_bucket, lsh_probes, lsh_multi_probe |
| Quantization | 4 | quantize_linear, dequantize |
| Int8 Distance | 4 | euclidean_int8, cosine_int8 |
| Temporal | 14 | time_now, time_diff, time_decay |
| Math | 13 | abs, sqrt, pow, log, sin, cos |
| String | 7 | len, upper, lower, substr, concat |
| Scalar Min/Max | 2 | min_val, max_val |
See Functions Reference for complete documentation.
// Facts
+relation(value1, value2) // Insert persistent fact
relation(value1, value2) // Insert session fact
-relation(value1, value2) // Delete fact
// Rules
+head(X, Y) <- body(X, Z), other(Z, Y) // Persistent rule
head(X, Y) <- body(X, Z), other(Z, Y) // Session rule
// Negation
result(X) <- source(X), !excluded(X)
// Queries
?relation(X, Y), X > 10
// Aggregations
?count<X> <- relation(X, _)
?sum<V> : G <- data(G, V) // Group by G
// Computed values
result(X, Y) <- input(X), Y = X * 2 + 1
// Comments
// Single line comment
/* Multi-line
comment */
See Syntax Cheatsheet for more.