Runtime error due to failure to criticize ontology

[sashapustota]

Hi there!

Im running into the following error when trying to parse a JSON file:

INFO:ontocast.agent.chunk_text:Chunking the text
INFO:ontocast.tool.chunk.chunker:chunk lengths: [753]
INFO:ontocast.agent.check_chunks:Chunks (rem): 1, chunks proc: 0. Setting up current chunk
INFO:ontocast.agent.check_chunks:Chunk available, setting status to FAILED and proceeding to SELECT_ONTOLOGY
INFO:ontocast.agent.select_ontology:Selecting ontology
INFO:ontocast.util:Starting to execute Text to Ontology
INFO:ontocast.agent.render_ontology_triples:Starting to render ontology triples
INFO:ontocast.agent.render_ontology_triples:Creating fresh ontology
INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
WARNING:ontocast.onto:Ontology IRI 'http://localhost:8999/ont_ltpqo' does not match expected 'http://localhost:8999/ltpqo'
INFO:ontocast.agent.render_ontology_triples:Ontology addendum has 54 triples.
INFO:ontocast.util:For Text to Ontology: status is SUCCESS, proceeding to next node
INFO:ontocast.util:Starting to execute Criticise Ontology
INFO:ontocast.agent.criticise_ontology:Criticize ontology
INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
INFO:ontocast.agent.criticise_ontology:Updating existing ontology: _void_ontology_name
ERROR:ontocast.cli.serve:Error processing document: 
ERROR:ontocast.cli.serve:Error type: <class 'RuntimeError'>
ERROR:ontocast.cli.serve:Error traceback:
Traceback (most recent call last):
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langchain_core/runnables/config.py", line 607, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/work/ontocastjobs/ontocast/ontocast/stategraph/util.py", line 51, in wrapper
    state = func(state)
            ^^^^^^^^^^^
  File "/work/ontocastjobs/ontocast/ontocast/agent/criticise_ontology.py", line 90, in criticise_ontology
    om_tool.update_ontology(
  File "/work/ontocastjobs/ontocast/ontocast/tool/ontology_manager.py", line 43, in update_ontology
    current_idx = next(
                  ^^^^^
StopIteration

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/work/ontocastjobs/ontocast/ontocast/cli/serve.py", line 176, in process
    async for chunk in workflow.astream(
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langgraph/pregel/__init__.py", line 1477, in astream
    async for _ in runner.atick(
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langgraph/pregel/runner.py", line 130, in atick
    await arun_with_retry(t, retry_policy, stream=self.use_astream)
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langgraph/pregel/retry.py", line 102, in arun_with_retry
    await task.proc.ainvoke(task.input, config)
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langgraph/utils/runnable.py", line 453, in ainvoke
    input = await asyncio.create_task(coro, context=context)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langgraph/utils/runnable.py", line 236, in ainvoke
    ret = await asyncio.create_task(coro, context=context)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langchain_core/runnables/config.py", line 616, in run_in_executor
    return await asyncio.get_running_loop().run_in_executor(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/concurrent/futures/thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/work/ontocastjobs/ontocast/.venv/lib/python3.12/site-packages/langchain_core/runnables/config.py", line 612, in wrapper
    raise RuntimeError from exc
RuntimeError

I have also tried disabling the ontology critique by adding SKIP_ONTOLOGY_DEVELOPMENT=true to my .env file, but that doesn't seem to help.

I am running the code from within the cloned repo, using the default uv run serve --ontology-directory ../data/ontologies --working-directory ../data and curl -X POST http://localhost:8999/process -F "file=@data/json/random.json"

Hope this is a right place to post such a problem. Thank you in advance.

[alexander-belikov]

Thank you for the feedback!

As we are preparing v0.1.7 release, you may check out the latest update on tag v0.1.7 or branch https://github.com/growgraph/ontocast/tree/18-add-memorycontext

There will be more than a few changes in .env but this specific issue should be solved.

Or you may wait for the official release while we are updating the docs.

[alexander-belikov]

@sashapustota please check - with the current release 0.2.3 this error should be taken care of

[sashapustota]

Hey @alexander-belikov. Thanks! Actually I can't seem to be able to spin up the server when I run:

 ontocast --env-path .env 

I get the following error:

Traceback (most recent call last):
File "/home/ucloud/.local/bin/ontocast", line 5, in
from ontocast.cli.serve import run
File "/work/ontosomething/ontocast/ontocast/cli/serve.py", line 41, in
from ontocast.cli.util import crawl_directories
File "/work/ontosomething/ontocast/ontocast/cli/util.py", line 4, in
from ontocast.tool.converter import ConverterTool
File "/work/ontosomething/ontocast/ontocast/tool/init.py", line 29, in
from .llm import LLMTool
File "/work/ontosomething/ontocast/ontocast/tool/llm.py", line 45, in
from langchain.output_parsers import PydanticOutputParser
ModuleNotFoundError: No module named 'langchain.output_parsers'

I am running it from within a cloned repo, which I install via pip install -e .

Any thoughts?

Thanks, as always!

[alexander-belikov]

@sashapustota ,

if you want to work with ontocast in development mode, please use uv
i.e. uv sync --all-groups

Otherwise you can run do pip install ontocast

[sashapustota]

@alexander-belikov,

The error is still there even if I just use pip install ontocast.

Anyways, the error is simply due to an incorrect import in the /tool/llm.py file, specifically from langchain.output_parsers import PydanticOutputParser should be from langchain**_core**.output_parsers import PydanticOutputParser; the code runs fine after fixing that.

However, now when I try to process a document, specifically json/chem.bassani-et-al-2024-nanocrystal-assemblies-current-advances-and-open-problems.json, I instantly get the following response in my terminal:

{"status":"success","data":{"facts":"","ontology":"@prefix owl: <http://www.w3.org/2002/07/owl#> .\n\n<https://example.com/__null__> a owl:Ontology .\n\n"},"metadata":{"status":"failed","chunks_processed":0,"chunks_remaining":29,"budget":{"chars_sent":0,"chars_received":0,"calls_count":0,"ontology_triples_generated":0,"facts_triples_generated":0,"ontology_operations_count":0,"facts_operations_count":0}}}

And the server terminal looks like this:

2025-12-19 05:26:36 - INFO - robyn.logger - info : Starting server at http://127.0.0.1:8999
2025-12-19 05:26:36 - INFO - actix_server.builder - None : starting 1 workers
2025-12-19 05:26:36 - INFO - actix_server.server - None : Actix runtime found; starting in Actix runtime
2025-12-19 05:26:36 - INFO - actix_server.server - None : starting service: "actix-web-service-127.0.0.1:8999", workers: 1, listening on: 127.0.0.1:8999
a
2025-12-19 05:29:17 - INFO - ontocast.agent.chunk_text - chunk_text : Chunking the text
2025-12-19 05:29:17 - INFO - ontocast.agent.chunk_text - chunk_text : Created 1 chunks for processing
2025-12-19 05:29:17 - INFO - ontocast.agent.select_ontology - select_ontology : Selecting ontology for chunk 1/1
2025-12-19 05:31:00 - INFO - ontocast.agent.chunk_text - chunk_text : Chunking the text
2025-12-19 05:31:00 - INFO - ontocast.agent.chunk_text - chunk_text : Created 30 chunks for processing
2025-12-19 05:31:00 - INFO - ontocast.agent.select_ontology - select_ontology : Selecting ontology for chunk 1/30

Any advice?