repoGitHubTrust 82 · PrimaryPublished 1mo agoLive · 5d ago
RimantasZ/contextspy
Context profiler for LLMs and AI agents - used to introspect context contents and reduce token costs
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 70%LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard →
- PossiblePossibly related (embedding) · 68%ACE: Pluggable Adaptive Context Elasticizer across Agents →
- PossiblePossibly related (embedding) · 55%ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning →
- PossiblePossibly related (embedding) · 54%ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping →
- PossiblePossibly related (embedding) · 54%I built an open-source memory governance layer for AI assistants - looking for technical feedback [P] →
- PossiblePossibly related (embedding) · 61%AI agents get smarter with context engineering - SiliconANGLE →
- PossiblePossibly related (embedding) · 64%AI agents get smarter with context engineering - siliconangle.com →
- PossiblePossibly related (embedding) · 51%CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents →
Implements
paperLLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive DashboardpaperACE: Pluggable Adaptive Context Elasticizer across AgentspaperReContext: Recursive Evidence Replay as LLM Harness for Long-Context ReasoningpaperShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping
Covers
Covers (incoming)
newsAI agents get smarter with context engineering - SiliconANGLEnewsAI agents get smarter with context engineering - siliconangle.comnewsLong Context Isn’t Free — I Built a Safe Prompt-Pruning Layer That Makes LLM Systems Work - Towards Data SciencenewsThe AI context gap: Enterprise AI organizations have a trust problem, not a retrieval problem — and most are still building the fixnewsShow HN: Ratel, give agents unlimited tools and skills without context bloatnewsVerifiers v1 Lets Agentic RL Training Exceed Model Context Windows via DAG Branching - Tech Times
Implements (incoming)
paperCompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon AgentspaperScoped Verification for Reliable Long-Horizon Agentic Context Evolution under Distribution ShiftpaperSelf-Guided Test-Time Training for Long-Context LLMspaperLongCrafter: Towards Diverse Long-Context Understanding via Evidence-Graph-Guided Instruction SynthesispaperDemonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at ScalepaperExtending LLM Context via Associative Recurrent Memory
Related across the graph
paperReContext: Recursive Evidence Replay as LLM Harness for Long-Context ReasoningnewsI built an open-source memory governance layer for AI assistants - looking for technical feedback [P]paperLongCrafter: Towards Diverse Long-Context Understanding via Evidence-Graph-Guided Instruction SynthesispaperExtending LLM Context via Associative Recurrent MemorypaperScoped Verification for Reliable Long-Horizon Agentic Context Evolution under Distribution ShiftnewsVerifiers v1 Lets Agentic RL Training Exceed Model Context Windows via DAG Branching - Tech TimespaperACE: Pluggable Adaptive Context Elasticizer across AgentsnewsThe AI context gap: Enterprise AI organizations have a trust problem, not a retrieval problem — and most are still building the fixpaperLLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive DashboardnewsAI agents get smarter with context engineering - SiliconANGLEpaperDemonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at ScalepaperShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic ShoppingnewsAI agents get smarter with context engineering - siliconangle.compaperCompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon AgentsnewsLong Context Isn’t Free — I Built a Safe Prompt-Pruning Layer That Makes LLM Systems Work - Towards Data SciencenewsShow HN: Ratel, give agents unlimited tools and skills without context bloatpaperSelf-Guided Test-Time Training for Long-Context LLMs