Add adversarial safety fixtures for deliberate reasoning engine (mcp server transforms linear reasoning)

[haasonsaas]

Summary

Exercise prompt/tool/data poisoning and fail-closed behavior for the repo's most sensitive agent-facing path.

This issue was generated from an org-wide EvalOps mining pass on 2026-05-10 07:57 UTC. It combines live GitHub repo signals with a per-repo arXiv search. Treat the research links as grounding for a concrete implementation, not as a request for a literature review.

Repo Evidence

• Repository description: MCP server that transforms linear AI reasoning into structured, auditable thought graphs
• Tree signals: 0 docs files, 1 workflows, 0 proto files, 1 test-like files.
• README.md:15 includes latent-spec language: - 🚨 Assumption Tracking: Monitor and invalidate assumptions with automatic cascade to dependent thoughts - 📊 Hypothesis Scoring: Track supporting and contradicting evidence (coming soon) - 💾 Session Persistence: Save and load reasoning sessions (coming soon)
• README.md:16 includes latent-spec language: - 📊 Hypothesis Scoring: Track supporting and contradicting evidence (coming soon) - 💾 Session Persistence: Save and load reasoning sessions (coming soon) - ✅ Graph Validation: Detect cycles, contradictions, and orphaned thoughts
• README.md:110 includes latent-spec language: const objective = await use_mcp_tool("dre", "log_thought", { thought: "Should we acquire Company X?", thought_type: "objective"
• examples/business-decision.md:6 includes latent-spec language: ## Scenario Your company is considering acquiring a competitor. You need to analyze whether this acquisition makes strategic sense.
• examples/business-decision.md:13 includes latent-spec language: { "thought": "Should we acquire TechCorp to expand our market position?", "thought_type": "objective"
• examples/business-decision.md:94 includes latent-spec language: ### 7. Invalidate Assumptions if Needed ```javascript

Research Grounding

Repo axes: memory, governance, evaluation, tooling

Search keywords: reasoning, https, npm, deliberate-reasoning-engine, dre, thought, github, thoughts, assumption, dependencies, run, claude

• arXiv:2504.08893v1 Knowledge Graph-extended Retrieval Augmented Generation for Question Answering (Jasper Linders, Jakub M. Tomczak), 2025.
• arXiv:2502.01113v3 GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation (Linhao Luo, Zicheng Zhao, Gholamreza Haffari, Dinh Phung, Chen Gong, Shirui Pan), 2025.
• arXiv:2504.05163v2 Evaluating Knowledge Graph Based Retrieval Augmented Generation Methods under Knowledge Incompleteness (Dongzhuoran Zhou, Yuqicheng Zhu, Xiaxia Wang, Yuan He, Jiaoyan Chen, Steffen Staab), 2025.
• arXiv:2508.09460v1 Towards Self-cognitive Exploration: Metacognitive Knowledge Graph Retrieval Augmented Generation (Xujie Yuan, Shimin Di, Jielong Tang, Libin Zheng, Jian Yin), 2025.
• arXiv:2512.20626v2 MegaRAG: Multimodal Knowledge Graph-Based Retrieval Augmented Generation (Chi-Hsiang Hsiao, Yi-Cheng Wang, Tzung-Sheng Lin, Yi-Ren Yeh, Chu-Song Chen), 2025.
• arXiv:2502.06864v1 Knowledge Graph-Guided Retrieval Augmented Generation (Xiangrong Zhu, Yuexiang Xie, Yi Liu, Yaliang Li, Wei Hu), 2025.
• arXiv:2506.21556v3 VAT-KG: Knowledge-Intensive Multimodal Knowledge Graph Dataset for Retrieval-Augmented Generation (Hyeongcheol Park, Jiyoung Seo, MinHyuk Jang, Hogun Park, Ha Dam Baek, Gyusam Chang), 2025.
• arXiv:2507.16826v1 A Query-Aware Multi-Path Knowledge Graph Fusion Approach for Enhancing Retrieval-Augmented Generation in Large Language Models (Qikai Wei, Huansheng Ning, Chunlong Han, Jianguo Ding), 2025.
• arXiv:2405.15436v1 Hybrid Context Retrieval Augmented Generation Pipeline: LLM-Augmented Knowledge Graphs and Vector Database for Accreditation Reporting Assistance (Candace Edwards), 2024.
• arXiv:2511.11017v1 AI Agent-Driven Framework for Automated Product Knowledge Graph Construction in E-Commerce (Dimitar Peshevski, Riste Stojanov, Dimitar Trajanov), 2025.

What To Build

• Add adversarial fixtures for prompt/tool/memory poisoning.
• Document the intended fail-closed behavior and any allowed degraded-mode fallback.
• Add regression coverage that proves unsafe inputs do not silently reach the privileged path.

Acceptance Criteria

• A short design note names the repo-specific workflow, threat or correctness model, and the research assumptions being adopted.
• A runnable check, fixture, or verifier exercises the new contract in CI or an equivalent local command documented in the repo.
• The implementation emits or stores enough evidence for a downstream agent/operator to cite inputs, decisions, and outputs.
• At least one negative/degraded-mode case is covered so failures are observable rather than silently accepted.
• Documentation links the new behavior to the relevant EvalOps platform primitive or explicitly records why this repo remains standalone.

Notes

• Generated issue 3/5 for evalops/deliberate-reasoning-engine by evalops_org_miner.py.
• Before implementation, confirm the sampled latent-spec snippets still match main; this issue intentionally cites exact file paths/lines where the mining pass saw them.