HKUDS/DeepTutor · error · ValueError

ConceptDesignAgent prompts are not configured.

Error message

ConceptDesignAgent prompts are not configured.

What it means

Catch-all in _extract_pdf: any unexpected exception from pypdf while reading pages is wrapped in CorruptDocumentError with the underlying message appended. It signals a malformed or unreadable PDF rather than a specific known condition.

Source

Thrown at deeptutor/agents/math_animator/agents/concept_design_agent.py:42

            agent_name="concept_design_agent",
            api_key=api_key,
            base_url=base_url,
            api_version=api_version,
            language=language,
        )

    async def process(
        self,
        *,
        user_input: str,
        output_mode: str,
        analysis: ConceptAnalysis,
        style_hint: str,
    ) -> SceneDesign:
        system_prompt = self.get_prompt("system")
        user_template = self.get_prompt("user_template")
        if not system_prompt or not user_template:
            raise ValueError("ConceptDesignAgent prompts are not configured.")

        user_prompt = user_template.format(
            user_input=user_input.strip(),
            output_mode=output_mode,
            style_hint=style_hint.strip() or "(none)",
            analysis_json=json.dumps(analysis.model_dump(), ensure_ascii=False, indent=2),
        )
        _chunks: list[str] = []
        async for _c in self.stream_llm(
            user_prompt=user_prompt,
            system_prompt=system_prompt,
            response_format={"type": "json_object"},
            stage="concept_design",
            trace_meta=build_trace_metadata(
                call_id=new_call_id("math-design"),
                phase="concept_design",
                label="Concept design",
                call_kind="math_concept_design",

View on GitHub (pinned to 3e82f13042)

Solutions

1. Repair the PDF first: qpdf --decrypt or Ghostscript can rebuild broken xref tables
2. Re-download/re-upload the file if it was truncated
3. Install pymupdf — fitz is tried first and tolerates many malformed PDFs pypdf rejects

Example fix

// before
text = extract_text_from_bytes(corrupted, filename="bad.pdf")

// after
import subprocess
tmp = "/tmp/bad.pdf"; open(tmp,"wb").write(corrupted)
subprocess.run(["qpdf", "--decrypt", tmp, "/tmp/fixed.pdf"], check=True)
text = extract_text_from_bytes(open("/tmp/fixed.pdf","rb").read(), filename="bad.pdf")

import fitz
try:
    with fitz.open(stream=data, filetype="pdf") as doc:
        doc.page_count  # forces full open; fitz tolerates many broken PDFs
except Exception:
    repair_or_reject(fn)

except CorruptDocumentError as e:
    if "failed to read PDF" in str(e):
        repaired = qpdf_repair(data)
        if repaired:
            text = extract_text_from_bytes(repaired, filename=fn)

When it happens

Trigger: pypdf throws ValueError ('stream has ended unexpectedly'), structural errors on truncated PDFs, or AssertionError on malformed xref tables while enumerating reader.pages.

Common situations: Truncated uploads (interrupted transfers), corrupted email attachments, hand-crafted or PDFs written by buggy generators.

Related errors

• CodeGeneratorAgent retry prompts are not configured.
• The model provider interrupted this response. Please retry.
• Unable to reach the model provider. Please retry.
• CodeGeneratorAgent generation prompts are not configured.
• ConceptAnalysisAgent prompts are not configured.

AI-assisted analysis of HKUDS/DeepTutor@3e82f13042 (2026-08-27). Data as JSON: /api/errors/9bec38cc4542898a. Report an issue: GitHub.