Clone a bank (async)
Changed onCopy this bank into a new one, in a single call. The clone starts with the source's memories as they are at clone time and evolves independently from then on: later retains, consolidation and edits on either bank leave the other alone.
This is the export and import above run back to back on this instance, so nothing is re-extracted and no LLM is called — facts are re-embedded and entities re-resolved, exactly as a restore does. The same three flags choose what the clone inherits: include_data (documents, facts, observations, attachments, the curation archive, the operations log, and the mental models and knowledge pages synthesized from them), include_bank_config (the bank's config overrides, directives and webhooks) and include_history (audit_log, llm_requests).
Note the webhooks: they travel with the bank's configuration, so a clone made with the default flags will call the source's webhook endpoints. Pass include_bank_config=false, or delete them on the clone, when they point at a per-bank consumer.
target_bank_id must not already exist. Returns an operation_id, recorded against the source bank (the target does not exist yet); poll GET /v1/default/banks/{bank_id}/operations/{operation_id} for status and the per-component counts.
Request
- Base URLhttps://hindsight.vectorize.io (the document declares no server; this is the origin it was published from)
- Authnone declared
- TypeScriptA function with typed inputs. Its response is checked against the schema.
Path parameters
Query parameters
Bank to create; must not already exist
Copy the memories, what backs them, and the mental models and knowledge pages synthesized from them
Copy the bank's config overrides, directives and webhooks
Copy audit_log and llm_requests
Headers
Response
Successful Response
Changes
- info
endpoint added
- info