Export format
An export is one archive of everything your tenant holds: your documents and their text, the passages and pages read from them, your entities, relationships and memory, your schema and its history, and your audit log. Your administrators request it from the console's Export your data page. It is ready once the archive is written, and it can be downloaded for seven days. While a deletion of your tenant can still be cancelled, an export made before it can still be downloaded.
This page is generated from a snapshot of what the export code writes, and a test fails if the two differ. Every field of every file is listed below.
The archive
The archive is a POSIX tar file (not compressed). Its first member is manifest.json; the other
files follow in the order listed below, each exactly once, even when it holds no records.
Every .jsonl file is JSON Lines: one JSON object per line, in UTF-8, each line ending with a line
feed. Each object's keys are in the canonical order of RFC 8785 (JSON Canonicalization Scheme), so
the same data always produces the same bytes. manifest.json is one canonical JSON object.
anchor.json is one canonical JSON object, or null when the audit log has no anchor yet.
All files are read from one consistent snapshot of your tenant's data, taken when the export ran: a record changed after that moment is not in the export, and every record in it is as it was at that moment.
Checking an export
manifest.json lists every file with its number of records, its size in bytes and its SHA-256
digest. To check an archive, extract it and compare each file's SHA-256 (for example with
sha256sum) and size with the manifest's. A file that is missing, extra, or differs is a damaged
archive: request a new export.
audit.jsonl holds your audit log up to the manifest's snapshotPosition, in position order, so it
can be read as the history of every change in the export. anchor.json holds the newest anchor of
that log, or null when the log has none yet.
How values are written
Each line of a .jsonl file is one stored record with every one of its fields, so a field can be
null only where the tables below say it can. The JSON column says how each value is written:
| JSON | Written as |
|---|---|
| string | A JSON string. |
| integer, number | A JSON number. |
| boolean | true or false. |
| timestamp | A string in ISO 8601 with a UTC offset, such as 2026-11-14T09:30:00.123456+00:00. |
| date | A string such as 2026-11-14. |
| json | The JSON value itself (an object or an array), not a string holding it. |
| array | A JSON array. |
| hex | A string of \x followed by the bytes in lowercase hexadecimal. |
| base64 | A string of standard base64 (RFC 4648), without line breaks. |
| base64url | A string of URL-safe base64 (RFC 4648 §5), without padding. |
The Stored as column names the database type the value comes from, for readers who load an export into their own database.
manifest.json
Format kg-export/1. The first file of the archive; it names every other file.
| Field | Meaning |
|---|---|
format | The format's name and version, kg-export/1. |
tenant | Your tenant's identifier. |
exportId | The export's identifier, as the console shows it. |
snapshotAt | When the snapshot the export was read from was taken (ISO 8601, UTC). |
snapshotPosition | The audit position the snapshot holds the log through: audit.jsonl ends at this position (0 when the log is empty). |
files | One entry for every other file, in archive order. |
| Field | Meaning |
|---|---|
name | The file's name in the archive. |
records | How many records the file holds: its lines, or for anchor.json 1 (0 when it holds null). |
bytes | The file's size in bytes. |
sha256 | The SHA-256 digest of the file's bytes, in lowercase hexadecimal. |
entity-types.jsonl
Your entity types, one per line, ordered by name.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
storage_id | string | no | text | The type definition's internal identifier. |
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
name | string | no | text | The type's name. |
identifying_property | string | no | text | The property whose value identifies an entity of this type. |
properties_json | json | no | text | The type's property definitions, with each property's status and constraints. |
created_seq | integer | no | bigint | A number that increases with each type created, giving creation order. |
status | string | no | text | Whether the type is active or deprecated. |
changed_revision | integer | no | bigint | The schema revision of the type's last change (0 for a type older than schema history). |
relationship-types.jsonl
Your relationship types, one per line, ordered by name.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
storage_id | string | no | text | The type definition's internal identifier. |
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
name | string | no | text | The type's name. |
allowed_source_types | array | no | ARRAY | The entity types a relationship of this type may start from. |
allowed_target_types | array | no | ARRAY | The entity types a relationship of this type may point to. |
properties_json | json | no | text | The type's property definitions, with each property's status and constraints. |
created_seq | integer | no | bigint | A number that increases with each type created, giving creation order. |
status | string | no | text | Whether the type is active or deprecated. |
changed_revision | integer | no | bigint | The schema revision of the type's last change (0 for a type older than schema history). |
schema-history.jsonl
Every applied schema change, one per line, in revision order.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
revision | integer | no | bigint | The schema revision this change produced, counting from 1. |
committed_at | timestamp | no | timestamp with time zone | When the change was applied. |
principal_id | string | no | text | The identity that made the change. |
actor_kind | string | no | text | Who made the change: a human or an agent. |
agent_id | string | yes | text | The agent's identifier when an agent made the change; null otherwise. |
request_id | string | no | text | The identifier of the request that made the change. |
operations_json | json | no | text | The change's operations in order, each with the element's definition before and after (for a very large element, the changed fields and a digest). |
entities.jsonl
Every entity, one per line, ordered by identifier. Documents, passages and memory entries are entities too, so they appear here as well as in their own files.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
id | string | no | text | The entity's identifier. |
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
type | string | no | text | The entity's type. |
identifying_value | string | yes | text | The value of its type's identifying property, when it has one. |
properties_json | json | no | text | The entity's properties. |
created_seq | integer | no | bigint | A number that increases with each entity created, giving creation order. |
identifying_digest | hex | yes | bytea | The SHA-256 of the identifying value, kept for lookups; null when there is no identifying value. |
identifying_normalized | string | yes | text | The identifying value with whitespace collapsed, compatibility characters folded (NFKC) and ASCII letters lower-cased, kept for duplicate suggestions; null when there is no identifying value, and briefly for entities stored before it existed. |
relationships.jsonl
Every relationship, one per line, ordered by identifier.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
id | string | no | text | The relationship's identifier. |
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
type | string | no | text | The relationship's type. |
source_id | string | no | text | The entity it starts from. |
target_id | string | no | text | The entity it points to. |
properties_json | json | no | text | The relationship's properties. |
created_seq | integer | no | bigint | A number that increases with each relationship created, giving creation order. |
extraction-receipts.jsonl
A receipt for every committed extraction, one per line: which reviewed candidates became entities and relationships. Extraction is off during the trial, so this file is normally empty.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
receipt_id | string | no | text | The receipt's identifier. |
preview_id | string | no | text | The extraction preview the commit approved. |
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
actor | string | no | text | The actor the preview was issued to. |
owner_key | string | no | text | The key of the caller who owned the preview; only that caller could commit it. |
principal_id | string | no | text | The identity that committed it. |
actor_kind | string | no | text | Who committed it: a human or an agent. |
approval_hash | string | no | text | A digest of the approval, used to recognise a repeated commit. |
review_hash | string | no | text | A digest of the reviewed preview. |
selection_hash | string | no | text | A digest of the selected candidates. |
receipt_json | json | no | text | The receipt: when it was committed, the source's digest, the model, and each change with its candidate, operation, record and evidence spans. It holds no source text. |
documents.jsonl
Every document, one per line, ordered by identifier.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
document_id | string | no | text | The document's identifier; it is part of every passage identifier. |
external_id | string | yes | text | Your own key for the document, if you gave one when uploading. |
entity_id | string | no | text | The Document entity that represents it in the graph. |
current_revision | integer | yes | integer | The revision in use; null until the first upload is committed. |
next_revision | integer | no | integer | The number the next upload will get. |
created_at | timestamp | no | timestamp with time zone | When the document was first uploaded. |
updated_at | timestamp | no | timestamp with time zone | When it last changed. |
revisions.jsonl
Every retained revision of every document, with its text, one per line, ordered by document and revision.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
document_id | string | no | text | The document. |
revision | integer | no | integer | The revision's number, counting from 1. |
state | string | no | text | Whether the revision is pending, current, superseded or discarded. |
media_type | string | no | text | The uploaded file's media type. |
original_digest | string | no | text | The SHA-256 of the file as uploaded, in lowercase hexadecimal. |
original_bytes | integer | no | integer | The size of the file as uploaded, in bytes. |
normalized_text | string | yes | text | The revision's text after normalization; passages are slices of it. Null for a discarded revision. |
normalized_digest | string | no | text | The SHA-256 of the normalized text, in lowercase hexadecimal. |
code_point_count | integer | no | integer | The normalized text's length in Unicode code points. |
normalization_rule | integer | no | integer | The version of the normalization rules that produced the text. |
splitting_rule | integer | no | integer | The version of the rules that split it into passages. |
passage_count | integer | yes | integer | How many passages the revision has; null until it is committed. |
uploaded_by | string | no | text | The identity that uploaded it. |
actor_kind | string | no | text | Who uploaded it: a human or an agent. |
agent_id | string | yes | text | The agent's identifier when an agent uploaded it; null otherwise. |
created_at | timestamp | no | timestamp with time zone | When it was uploaded. |
committed_at | timestamp | yes | timestamp with time zone | When it became current; null if it never did. |
superseded_at | timestamp | yes | timestamp with time zone | When a newer revision replaced it; null while it is current. |
page_total | integer | yes | integer | For a PDF, its number of pages; null otherwise. |
images_skipped | integer | yes | integer | How many images in the file were not read; null for a format that has none. |
complete | boolean | no | boolean | False while any page of a PDF failed to be read. |
no_text | boolean | no | boolean | True for a PDF whose pages were all blank, which has no text and no passages. |
passages.jsonl
Every passage of every committed revision, one per line, ordered by document, revision and position. A passage's text is not repeated: it is the slice of its revision's normalized_text from span_start to span_end.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
document_id | string | no | text | The document. |
revision | integer | no | integer | The revision. |
ordinal | integer | no | integer | The passage's position in the revision. |
span_start | integer | no | integer | Where the passage starts in the revision's normalized text, in code points from 0. |
span_end | integer | no | integer | Where it ends, in code points (exclusive). |
heading_path | string | no | text | The headings the passage sits under, in full. |
entity_id | string | yes | text | The Passage entity while the revision is current; null after it is superseded. |
page | integer | yes | integer | For a PDF, the page the passage starts on; null otherwise. |
originals.jsonl
Each uploaded PDF or Word file exactly as it was uploaded, one per line, ordered by document and revision. Its media type, SHA-256 and size are its revision's media_type, original_digest and original_bytes.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
document_id | string | no | text | The document. |
revision | integer | no | integer | The revision. |
content_base64 | base64 | no | bytea | The file's bytes. |
pages.jsonl
Every page of every PDF revision and what reading it produced, one per line, ordered by document, revision and page.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
document_id | string | no | text | The document. |
revision | integer | no | integer | The revision. |
page | integer | no | integer | The page's number, counting from 1. |
label | string | yes | text | The page's label in the PDF, if it has one. |
state | string | no | text | Whether the page is pending, read, empty or failed. |
reason | string | yes | text | Why the page failed; null unless it failed. |
text | string | yes | text | The page's text as read, in Markdown. |
model | string | yes | text | The model that read the page. |
settings_key | string | yes | text | The rendering settings the page was read with. |
instruction_version | integer | yes | integer | The version of the reading instruction used. |
tokens | integer | no | bigint | The tokens the model provider reported for reading the page, input and output together. |
reused | boolean | no | boolean | True when the page was copied from an earlier upload of the same file, read the same way, instead of being read again. |
audit.jsonl
Your audit log up to the manifest's snapshotPosition: every committed change, one per line, in position order.
| Field | JSON | Can be null | Stored as | Meaning |
|---|---|---|---|---|
tenant_id | string | no | text | Your tenant's identifier; the same on every record. |
position | integer | no | bigint | The entry's position in the log, consecutive from 1 in commit order. |
committed_at | timestamp | no | timestamp with time zone | When the change was committed. |
principal_id | string | no | text | The identity that made the change. |
actor_kind | string | no | text | Who made the change: a human, an agent, or the service itself (system). |
agent_id | string | yes | text | The agent's identifier when an agent made the change; null otherwise. |
request_id | string | no | text | The identifier of the request that made the change. |
operation | string | no | text | What was done, such as create, update or delete. |
record_kind | string | no | text | The kind of record changed, such as entity, relationship or schema. |
record_type | string | yes | text | The record's type, when it has one. |
record_id | string | yes | text | The record's identifier, when it has one. |
before_json | json | yes | text | The record's complete state before the change; null for a creation. |
after_json | json | yes | text | The record's complete state after the change; null for a deletion. |
refs_json | json | yes | text | Other records the change refers to, when there are any. |
confirmation | string | yes | text | For an operation that needed confirmation, whether it was confirmed. |
key_id | string | no | text | The identifier of the key the entry's MAC was computed with. |
mac | base64url | no | bytea | The entry's message authentication code, which chains it to the previous entry, so the service can detect a changed or missing entry. |
anchor.json
The newest anchor of your audit log at the snapshot: a record, kept outside your tenant's data, of a position in the log and the MAC of the entry there. It is null when the log has no anchor yet.
| Field | Meaning |
|---|---|
tenant | Your tenant's identifier. |
position | The audit position the anchor was taken at. |
entryMac | The MAC of the audit entry at that position, in URL-safe base64; it equals that entry's mac in audit.jsonl. |
anchoredAt | When the anchor was recorded (ISO 8601, UTC). |