Operator-ready prompt for reuse, tuning, and workspace runs.
This item is set up for developers who want to inspect the original language, fork it into Workspace, and adapt the evidence model without losing the source prompt structure.
Implementation handoffs, eval setup, and prompt tuning where you need the original structure intact.
Inspect first, copy once, then fork into Workspace when you want variants, notes, and model settings attached to the same run.
Swap domain facts, examples, and any hard-coded entities for your own context.
Tighten the evidence or verification requirement if this is headed toward production.
Decide which failure mode you want to evaluate first before you branch the prompt.
This prompt already carries implementation detail, tool context, and a final-output instruction. Keep that structure intact when you tune it, or your comparison runs get noisy fast.
Open this prompt inside Workspace when you want a live iteration loop.
Copy for quick reuse, or run it in Workspace to keep prompt variants, model settings, and prompt-history changes in one place.
Structured source with 49 active lines to adapt.
Already linked to a challenge workflow.
Sign in to keep private prompt variations.
Prompt content
Original prompt text with formatting preserved for inspection and clean copy.
Define a set of `Task` objects that guide the agents through the scientific review process. Create tasks for summarizing the input abstract (using a Mistral Saba based agent if possible), verifying facts, checking consistency, and detecting AI slop. Finally, define the `Crew` with these agents and tasks, specifying the `Process.sequential` flow. The output of the crew should be a comprehensive review report.
```python
# ... (agents defined previously)
# Summarizer Agent (using Mistral Saba for efficiency)
# mistral_saba_llm = ChatMistralAI(model="mistral-large-latest", temperature=0.1, mistral_api_key="YOUR_MISTRAL_API_KEY")
# summarizer_agent = Agent(role='Summarizer', goal='Condense complex texts into concise summaries.', llm=mistral_saba_llm, ...)
# Define Tasks
summarize_task = Task(
description="Summarize the provided scientific abstract into 3-4 key bullet points.",
agent=None, # Assign summarizer_agent here
expected_output="A bulleted list summarizing the abstract.",
output_file="summary.md"
)
verify_facts_task = Task(
description="Verify all factual claims in the abstract using the Pinecone Fact Checker tool. Identify and justify any inaccuracies.",
agent=factual_verifier,
expected_output="A list of factual claims with their verification status and justifications.",
output_file="factual_verification.md"
)
check_consistency_task = Task(
description="Analyze the abstract for logical consistency, internal contradictions, and methodological soundness. Highlight any areas needing clarification or revision.",
agent=consistency_checker,
expected_output="A report detailing inconsistencies found and suggestions for improvement.",
output_file="consistency_check.md"
)
detect_ai_slop_task = Task(
description="Examine the writing style for signs of generic, vague, or repetitive language typical of AI-generated content. Provide specific examples and a likelihood score.",
agent=ai_slop_detector,
expected_output="A report on potential 'AI slop' indicators with examples and an AI generation likelihood score.",
output_file="ai_slop_detection.md"
)
# Final Review Task (assigned to a lead agent or the consistency_checker for synthesis)
final_review_task = Task(
description="Synthesize findings from all agents into a final comprehensive review report, including an overall verdict and actionable recommendations.",
agent=consistency_checker, # Could be a new 'Review Synthesizer' agent
expected_output="A comprehensive review report combining all findings, with an overall verdict on scientific integrity and actionable recommendations.",
context=[summarize_task, verify_facts_task, check_consistency_task, detect_ai_slop_task]
)
# Form the Crew
scientific_review_crew = Crew(
agents=[factual_verifier, consistency_checker, ai_slop_detector], # Add summarizer_agent if created
tasks=[summarize_task, verify_facts_task, check_consistency_task, detect_ai_slop_task, final_review_task],
process=Process.sequential, # Or hierarchical for more complex flows
verbose=2 # Enables detailed logging
)
# To run the crew:
# result = scientific_review_crew.kickoff(inputs={'abstract': 'Your scientific abstract text here.'})
# print(result)
```Adaptation plan
Keep the source stable, then branch your edits in a predictable order so the next prompt run is easier to evaluate.
Preserve the role framing, objective, and reporting structure so comparison runs stay coherent.
Swap in your own domain constraints, anomaly thresholds, and examples before you branch variants.
Check whether the prompt asks for the right evidence, confidence signal, and escalation path.
Copy once for a pristine source snapshot, then move the prompt into Workspace when you want variants, run history, and side-by-side tuning without losing the original.
Prompt diagnostics
Quick signals for how structured this prompt already is and where adaptation work is likely to happen first.
This prompt already mixes executable detail with instructions, so the safest path is to tune examples and interfaces before you rewrite the overall scaffold.
Orchestrate Scientific Integrity Agent Crew
With growing concerns about 'AI slop' in scientific publishing, this challenge focuses on developing an agentic system to enforce scientific integrity. You will use CrewAI to orchestrate a team of specialized AI agents that act as a 'Scientific Review Board.' This crew will collaborate to analyze newly generated scientific abstracts or summaries, identify potential factual inaccuracies, inconsistencies, and characteristics of AI-generated content, and verify claims against a knowledge base. The system should highlight suspicious areas and provide justifications for its findings, leveraging the advanced reasoning capabilities of Claude Opus 4.1.
Use the challenge page to recover the original task boundaries before you tune the prompt. That keeps your variants grounded in the same evaluation target instead of drifting into a different problem.