SDK Function Reference

This page is the complete index of the prometheux_chain Python SDK. Every function is re-exported at the top level, so you call it as px.<function>(...):

import prometheux_chain as px

ontology_id = px.save_ontology(ontology_name="my_ontology")

Each function is a thin wrapper around a REST endpoint — follow the Endpoint link to the full reference (parameters, request body, and response shape). Path segments like {ontology_id} are supplied as function arguments.

A few functions are generators that stream events as they arrive (newline-delimited JSON or a WebSocket). They are flagged below — iterate over the return value rather than awaiting a single result. See Streaming responses.

Config

The config object reads and writes local SDK settings (token, backend URL, LLM keys). See Config for the full list of keys.

Function	Description
`config.get(key, default=None)`	Read a local config value.
`config.set(key, value)`	Set a local config value.
`config.update_config(updates)`	Set multiple config values at once.

Ontologies

Function	Endpoint	Description
`save_ontology(...)`	`POST /ontologies/save`	Create or update an ontology.
`list_ontologies(...)`	`GET /ontologies/list`	List every ontology visible to you.
`load_ontology(...)`	`GET /ontologies/load`	Load full details for one ontology.
`copy_ontology(...)`	`POST /ontologies/copy`	Duplicate an ontology.
`list_templates(...)`	`GET /ontologies/list-templates`	Browse marketplace templates.
`import_template(...)`	`POST /ontologies/import-template`	Create an ontology from a template.
`export_ontology(...)`	`POST /ontologies/export-ontology`	Serialize an ontology to a JSON payload.
`import_ontology(...)`	`POST /ontologies/import-ontology`	Deprecated. Restore an ontology from a payload.
`export_workspace(...)`	`POST /ontologies/export-workspace`	Deprecated. Export every ontology in the workspace.
`import_workspace(...)`	`POST /ontologies/import-workspace`	Deprecated. Restore a full workspace.
`create_ontology_from_context(...)`	`POST /ontologies/create-from-context`	Build an ontology from text/files via the LLM.
`create_snapshot(...)`	`POST /ontologies/snapshots/create`	Capture a point-in-time snapshot.
`list_snapshots(...)`	`GET /ontologies/snapshots/list`	List an ontology’s snapshots.
`restore_snapshot(...)`	`POST /ontologies/snapshots/restore`	Restore an ontology from a snapshot.
`delete_snapshot(...)`	`POST /ontologies/snapshots/delete`	Delete a snapshot.
`cleanup_ontologies(...)`	`POST /ontologies/cleanup`	Delete one ontology or all.

Function	Endpoint	Description
`create_share(...)`	`POST /ontologies/share`	Share an ontology as viewer/editor.
`list_shares(...)`	`GET /ontologies/share/list`	List shares you created (outbox).
`update_share_role(...)`	`POST /ontologies/share/update-role`	Change a share between viewer/editor.
`revoke_share(...)`	`POST /ontologies/share/revoke`	Revoke a share you granted.
`list_inbox(...)`	`GET /ontologies/share/inbox`	List shares granted to you (inbox).
`accept_share(...)`	`POST /ontologies/share/accept`	Accept a pending share.
`leave_share(...)`	`POST /ontologies/share/leave`	Detach from an accepted share.
`sync_inbox(...)`	`POST /ontologies/share/sync`	Reconcile local shares with the source of truth.

Concepts

Function	Endpoint	Description
`save_concept(...)`	`POST /concepts/{ontology_id}/save`	Create or update a concept.
`list_concepts(...)`	`GET /concepts/{ontology_id}/list`	List concepts in a project.
`run_concept(...)`	`POST /concepts/{ontology_id}/run/{concept_name}`	Run a concept and its dependency chain.
`run_concept_stream(...)`	`WS /concepts/{ontology_id}/run-stream/{concept_name}`	Run a concept, streaming progress frames.
`rename_concept(...)`	`POST /concepts/{ontology_id}/rename`	Rename a concept and its dependents.
`reorder_concepts(...)`	`POST /concepts/{ontology_id}/reorder`	Set concept display order.
`fetch_results(...)`	`POST /concepts/{ontology_id}/fetch`	Read persisted output rows (paginated, filterable).
`search_results(...)`	`POST /concepts/{ontology_id}/fetch`	Convenience wrapper over `fetch_results` for term/column filtering.
`query_concept(...)`	`POST /concepts/{ontology_id}/query`	Read-only SQL `SELECT` over a populated concept.
`search_similar_concepts(...)`	`GET /concepts/search-similar`	Semantic search for concepts across projects.
`get_execution_status(...)`	`GET /concepts/{ontology_id}/execution-status`	Latest run status for one project.
`get_execution_statuses(...)`	`GET /concepts/execution-statuses`	Run statuses across all projects.
`generate_concept_description(...)`	`POST /concepts/{ontology_id}/generate-description`	LLM-generate a concept description.
`get_concept_description(...)`	`GET /concepts/{ontology_id}/description`	Read/lazily-generate a description.
`llm_analysis(...)`	`POST /concepts/{ontology_id}/llm`	Ask a question over predicate data.
`download_concept(...)`	`GET /concepts/{ontology_id}/download`	Export results to CSV / download a file.
`cleanup_concepts(...)`	`POST /concepts/{ontology_id}/cleanup`	Delete concepts.

Data sources & files

Function	Endpoint	Description
`connect_sources(...)`	`POST /data/connect`	Register a database/file source.
`list_sources(...)`	`GET /data/list`	List registered sources.
`list_sheets(...)`	`POST /data/list-sheets`	List sheets in a spreadsheet source.
`list_demo_sources(...)`	`GET /data/demo-sources`	List onboarding demo sources.
`refresh_sources(...)`	`POST /data/refresh`	Re-probe and reconcile sources.
`infer_schema(...)`	`POST /data/infer-schema`	Generate Vadalog binds from a connection.
`preview_datasource(...)`	`POST /data/preview`	Preview rows from a bound source.
`all_pairs_join(...)`	`POST /data/all-pairs-join`	Discover joins across sources.
`upload_file(...)`	`POST /data/files/upload`	Upload a file to `disk/`.
`list_files(...)`	`GET /data/files/list`	List files in `disk/`.
`make_directory(...)`	`POST /data/files/mkdir`	Create a directory in `disk/`.
`move_file(...)`	`POST /data/files/move`	Move/rename a file or directory.
`delete_files(...)`	`POST /data/files/delete`	Delete files/directories.
`download_file(...)`	`GET /data/files/download`	Download a file from `disk/`.
`cleanup_sources(...)`	`POST /data/cleanup`	Remove data sources.

Schema

Function	Endpoint	Description
`save_ontology_schema(...)`	`POST /concepts/{ontology_id}/save-ontology-schema`	Persist an ontology’s schema.
`load_ontology_schema(...)`	`GET /concepts/{ontology_id}/load-ontology-schema`	Load an ontology’s schema.
`update_concept_ontology_schema_type(...)`	`PATCH /concepts/{ontology_id}/concepts/{concept_name}/ontology-schema-type`	Mark a concept as node/edge.
`add_to_lineage(...)`	`POST /concepts/{ontology_id}/add-to-lineage`	Add a node/edge to the lineage.
`import_owl(...)`	`POST /concepts/{ontology_id}/import-owl`	Import OWL into editor JSON.

Knowledge graph

Function	Endpoint	Description
`build_graph(...)`	`POST /kgs/{ontology_id}/build-graph/{output_predicate}`	Build a graph from an edge predicate.
`run_graph_analytics(...)`	`POST /kgs/{ontology_id}/graph-analytics/{output_predicate}`	Run an analytics algorithm on a graph.
`list_graph_functions(...)`	`GET /kgs/graph-functions`	List available analytics functions.
`visualize_concept_lineage(...)`	`POST /kgs/{ontology_id}/visualize-concept-lineage`	Return the concept dependency DAG.

Context layer (knowledge)

Function	Endpoint	Description
`list_context_notes(...)`	`GET /knowledge/context`	List context notes.
`create_context_note(...)`	`POST /knowledge/context`	Create a context note.
`get_context_note(...)`	`GET /knowledge/context/{note_id}`	Read a context note.
`update_context_note(...)`	`PATCH /knowledge/context/{note_id}`	Update a context note.
`delete_context_note(...)`	`DELETE /knowledge/context/{note_id}`	Delete a context note.
`search_context_notes(...)`	`POST /knowledge/context/search`	Semantic search over notes.
`create_context_notes_from_file(...)`	`POST /knowledge/context/from-file`	Ingest a file as context notes.
`create_context_edge(...)`	`POST /knowledge/context/edges`	Link a note to a concept or ontology.
`auto_seed(...)`	`POST /knowledge/auto-seed`	Profile sources, stream observations.
`ontology_text(...)`	`GET /knowledge/ontology/{ontology_id}/text`	Render the ontology document.
`interview_template(...)`	`GET /knowledge/interview/template`	Onboarding interview template.
`submit_interview(...)`	`POST /knowledge/interview`	Submit onboarding interview answers.
`onboarding_status(...)`	`GET /knowledge/onboarding-status`	Report onboarding progress.

AI agent

Function	Endpoint	Description
`agent_chat(...)`	`POST /agent/{ontology_id}/chat`	Stream a conversation with the agent.
`agent_reset(...)`	`POST /agent/{ontology_id}/reset`	Clear a session’s history.

Chat history

Function	Endpoint	Description
`list_sessions(...)`	`GET /chat-history/sessions`	List agent chat sessions.
`get_session(...)`	`GET /chat-history/sessions/{session_id}`	Load a session with messages.
`rename_session(...)`	`PATCH /chat-history/sessions/{session_id}`	Rename a session.
`delete_session(...)`	`DELETE /chat-history/sessions/{session_id}`	Delete a session.

Apps

Function	Endpoint	Description
`list_all_apps(...)`	`GET /apps/list`	List apps across ontologies.
`list_apps(...)`	`GET /apps/{ontology_id}/list`	List apps in an ontology.
`get_app(...)`	`GET /apps/{ontology_id}/{app_id}`	Load an app.
`save_app(...)`	`POST /apps/{ontology_id}/save`	Create or update an app.
`publish_app(...)`	`POST /apps/{ontology_id}/{app_id}/publish`	Publish a frozen snapshot of the draft.
`unpublish_app(...)`	`POST /apps/{ontology_id}/{app_id}/unpublish`	Remove the published snapshot.
`delete_app(...)`	`DELETE /apps/{ontology_id}/{app_id}`	Delete an app.

Schedules

Function	Endpoint	Description
`create_policy(...)`	`POST /schedules/{ontology_id}/policies`	Attach a cron/data-change policy.
`list_policies(...)`	`GET /schedules/{ontology_id}/policies`	List policies in a project.
`get_policy(...)`	`GET /schedules/{ontology_id}/policies/{policy_id}`	Load a policy.
`update_policy(...)`	`PATCH /schedules/{ontology_id}/policies/{policy_id}`	Update or enable/disable a policy.
`delete_policy(...)`	`DELETE /schedules/{ontology_id}/policies/{policy_id}`	Delete a policy.
`trigger_policy(...)`	`POST /schedules/{ontology_id}/policies/{policy_id}/trigger`	Fire a policy run immediately.
`get_run_history(...)`	`GET /schedules/{ontology_id}/policies/{policy_id}/runs`	Read a policy’s run history.

Alerts

Function	Endpoint	Description
`get_alert_history(...)`	`GET /alerts/history`	List stored alerts.
`reprocess_alert(...)`	`POST /alerts/{alert_id}/reprocess`	Re-run an alert’s concept.

Compute

Function	Endpoint	Description
`check_compute_availability(...)`	`POST /compute/availability`	Check machine/Databricks availability.
`list_machines_combined(...)`	`GET /machines/machines-combined`	The machine catalogue plus your own machines.
`use_machine(...)`	`POST /machines/user-machines/use`	Add a catalogue machine to your list. Adding does not start it, so nothing is billed yet.
`set_machine_active(...)`	`PATCH /machines/user-machines/toggle-active`	Start or stop a machine; starting opens a billable session.
`get_machine_status(...)`	`GET /machines/user-machines/{user_machine_id}/status`	Real-time status of one machine.
`delete_user_machine(...)`	`DELETE /machines/user-machines/{user_machine_id}`	Remove a machine from your list, keeping its usage history.
`set_engine_workers(replicas, idle_minutes=None)`	`PUT /machines/workers`	Run `replicas` engine workers beside the running machine (0 removes them). Up to 4, each billed at the machine’s hourly rate; removed after `idle_minutes` (default 10, never longer than the machine’s auto-termination) without work.
`get_engine_workers()`	`GET /machines/workers`	How many workers are wanted and ready, how evaluations are spread across the engines, and what the workers cost (`billing`).
`wait_for_engine_workers(replicas, timeout=600)`	`GET /machines/workers`	Wait until `replicas` workers are ready; raises `TimeoutError` otherwise.
`engine_workers(replicas, ...)`	—	Context manager: start the workers, wait for them, and remove them on exit.

machine_id identifies a machine in the catalogue and is only used by use_machine; user_machine_id identifies one already on your list. Engine workers are extra copies of the engine next to your machine. Each evaluate_program / check_program call goes to the least busy engine, so a test suite that sends evaluations in parallel finishes sooner:

from concurrent.futures import ThreadPoolExecutor
import prometheux_chain as px

with px.engine_workers(2):
    with ThreadPoolExecutor(max_workers=3) as pool:
        reports = list(pool.map(lambda p: px.check_program(p, expect=expected[p]), programs))

Workers only take evaluations; concept runs and everything else stay on the machine. Each worker is a node the size of your machine and is billed at your machine’s hourly rate, from when you ask for it until it is removed. Worker time appears on your invoice under your machine, and free minutes don’t apply to it. Adding workers to a paid machine needs a payment method, just like starting the machine.

Assistant

Function	Endpoint	Description
`list_skills(...)`	`GET /assistant/skills`	List the assistant’s skill playbooks (metadata only).
`get_skill(...)`	`GET /assistant/skills/{skill_id}`	Fetch one skill’s full playbook.
`get_company_info(...)`	`GET /assistant/company-info`	Search the Prometheux company knowledge base.

Vadalog

Function	Endpoint	Description
`analyze_program(...)`	`POST /vadalog/analyze`	Inspect a program’s predicate structure.
`parse_binds(...)`	`POST /vadalog/parse-binds`	Separate rules from bind annotations.
`build_bind(...)`	`POST /vadalog/build-bind`	Rewrite a bind for a new predicate.
`generate_program(...)`	`POST /vadalog/program`	Write a Vadalog program from a natural-language question.
`evaluate_program(...)`	`POST /vadalog/evaluate`	Evaluate a Vadalog program directly.
`validate_concept(...)`	`POST /vadalog/validate`	Check a concept body before saving it.

Vadalingo

Function	Endpoint	Description
`translate_sql_to_vadalog(...)`	`POST /vadalingo/{ontology_id}/translate/sql-to-vadalog`	Translate SQL to Vadalog.
`translate_rdf_to_vadalog(...)`	`POST /vadalingo/{ontology_id}/translate/rdf-to-vadalog`	Translate RDF to Vadalog.
`translate_owl_to_vadalog(...)`	`POST /vadalingo/{ontology_id}/translate/owl-to-vadalog`	Translate OWL to Vadalog.

Testing

Unit tests for Vadalog programs: evaluate a program and compare each output predicate with the rows you expect. See Unit-test a Vadalog program.

Function	Endpoint	Description
`check_program(program, expect=None, mode="exact", ...)`	`POST /vadalog/evaluate`	Evaluate a program and check its output predicates; returns a `CheckReport`.
`check_question(question, expect=None, mode="exact", ...)`	`POST /vadalog/program`, then `POST /vadalog/evaluate`	Generate a program from a question, then check it like `check_program`.
`compare_rows(predicate, actual, expected, mode="exact")`	—	Compare two sets of rows locally; returns a `PredicateCheck`.
`read_expected_csv(path, header=False)`	—	Read expected rows from a CSV file.

Account & configuration

Function	Endpoint	Description
`get_role(...)`	`GET /users/get-role`	Get the caller’s role.
`get_usage_status(...)`	`GET /users/usage-status`	LLM/embedding usage and limits.
`get_login_activity(...)`	`GET /users/login-activity`	Recent access IPs/sessions.
`list_llm_models(...)`	`POST /users/llm-models`	List models for an LLM provider.
`save_user_config(...)`	`POST /users/save-config`	Persist server-side user config.
`load_user_config(...)`	`GET /users/load-config`	Load server-side user config.

Authentication

Function	Endpoint	Description
`issue_token(...)`	`POST /auth/issue-token`	Mint a new API key.
`list_tokens(...)`	`GET /auth/tokens`	List your API keys’ metadata.
`revoke_token(...)`	`POST /auth/revoke`	Revoke the current token.
`revoke_specific_token(...)`	`POST /auth/revoke/{jti}`	Revoke a token by ID.
`revoke_all_tokens(...)`	`POST /auth/revoke-all`	Revoke all your tokens.

Introduction

Get Started

Platform

Integrations

Examples

Vadalog Reference

Resources

SDK Function Reference

Config

Ontologies

Concepts

Data sources & files

Schema

Knowledge graph

Context layer (knowledge)

AI agent

Chat history

Apps

Schedules

Alerts

Compute

Assistant

Vadalog

Vadalingo

Testing

Account & configuration

Authentication

​Config

​Ontologies

​Ontology sharing

​Concepts

​Data sources & files

​Schema

​Knowledge graph

​Context layer (knowledge)

​AI agent

​Chat history

​Apps

​Schedules

​Alerts

​Compute

​Assistant

​Vadalog

​Vadalingo

​Testing

​Account & configuration

​Authentication

Config

Ontologies

Ontology sharing

Concepts

Data sources & files

Schema

Knowledge graph

Context layer (knowledge)

AI agent

Chat history

Apps

Schedules

Alerts

Compute

Assistant

Vadalog

Vadalingo

Testing

Account & configuration

Authentication