{ "schema_version": "1.1", "id": "atlas-retrieval", "slug": "retrieval", "title": "Retrieval", "url": "https://feed7.dev/atlas/retrieval", "current_answer": null, "implementation_consequence": null, "agent_context": null, "confidence": "auto_collected", "last_verified": null, "last_updated": null, "evidence": [ { "schema_version": "1.1", "id": "auto-e7ee7f4465", "slug": "lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465", "url": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465", "title": "Lessons from Studying Every Memory System — Shlok Khemani, Independent", "why_included": "Make agent memory inspectable and editable, with conflict detection and an explicit update cadence to prevent false beliefs from persisting.", "summary": "Persistent memory is a compute and product tradeoff, not just retrieval. Profiles need conflict detection, visibility, editing, and deliberate update cadence before agents can rely on them.", "practical_implication": "For agent memory, choose update frequency and profile size as an explicit compute budget. Make stored beliefs inspectable and editable, preserve source context where possible, and detect uncertainty or contradictions before a profile silently steers future work.", "agent_context": "ChatGPT and Claude have converged on running user profiles plus tools for retrieving past conversations. The observed implementations differ: ChatGPT's profile is about **4,000 tokens** and updates every few days, while Claude's is about **1,000 tokens** and updates every 24 hours.\n\nFor agent memory, choose update frequency and profile size as an explicit compute budget. Make stored beliefs inspectable and editable, preserve source context where possible, and detect uncertainty or contradictions before a profile silently steers future work.\n\nThe Turkey example shows the core failure: conversations about possible travel were condensed into a trip that never happened. Neither profile summaries nor retrieval alone solve missing evidence, cross-product context silos, or reconciliation with email, calendars, and other sources.", "source": { "name": "AI Engineer", "url": "https://www.youtube.com/watch?v=5ZGyKWjQDr0", "published_at": "2026-08-12T00:00:00.000Z" }, "source_class": "video", "content_type": "Video", "layer": "agent", "domains": [], "topics": [ "agent-memory", "context-engineering", "retrieval" ], "verification": { "status": "source_linked", "label": "Source Linked", "method": "source_feed", "verified_at": null }, "uncertainty": [], "connected_context": null, "lifecycle": "New", "published_at": "2026-08-12T00:00:00.000Z", "modified_at": "2026-08-12T00:00:00.000Z", "supersedes": [], "expires_at": null, "formats": { "html": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465", "json": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465.json", "markdown": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465.md" } }, { "schema_version": "1.1", "id": "auto-3c0b2ed056", "slug": "building-agents-is-trivial-now-context-is-the-next-front-3c0b2ed056", "url": "https://feed7.dev/p/building-agents-is-trivial-now-context-is-the-next-front-3c0b2ed056", "title": "Building Agents Is Trivial Now, Context Is the Next Frontier — Jeff Ng, Unblocked", "why_included": "Retrieve code alongside decisions, incidents, tickets, and conversations so the agent does not repeat a known outage trigger.", "summary": "A coding agent with repository access still recommended a known outage trigger. Background agents need reconciled decisions and incident context, not just more raw MCP results.", "practical_implication": "For unattended coding agents, retrieve decisions, incidents, tickets, conversations, and code together. Reconcile conflicts, rank relevance, enforce permissions, and provide a synthesized task view rather than making each agent resolve raw MCP output.", "agent_context": "An agent investigating QA latency of **3–4 seconds** versus an expected **hundreds of milliseconds** recommended re-enabling async dispatch. It missed a later Slack discussion and Linear postmortem showing that setting had caused an outage.\n\nFor unattended coding agents, retrieve decisions, incidents, tickets, conversations, and code together. Reconcile conflicts, rank relevance, enforce permissions, and provide a synthesized task view rather than making each agent resolve raw MCP output.\n\nThe example demonstrates one corrected recommendation, not a measured reliability gain. Building and maintaining an organization-wide context model also introduces access-control, freshness, and conflict-resolution work.", "source": { "name": "AI Engineer", "url": "https://www.youtube.com/watch?v=HvMyYLTfvhg", "published_at": "2026-08-21T00:00:00.000Z" }, "source_class": "video", "content_type": "Video", "layer": "context", "domains": [ "coding" ], "topics": [ "context-engineering", "retrieval", "mcp" ], "verification": { "status": "source_linked", "label": "Source Linked", "method": "source_feed", "verified_at": null }, "uncertainty": [], "connected_context": null, "lifecycle": "New", "published_at": "2026-08-21T00:00:00.000Z", "modified_at": "2026-08-21T00:00:00.000Z", "supersedes": [], "expires_at": null, "formats": { "html": "https://feed7.dev/p/building-agents-is-trivial-now-context-is-the-next-front-3c0b2ed056", "json": "https://feed7.dev/p/building-agents-is-trivial-now-context-is-the-next-front-3c0b2ed056.json", "markdown": "https://feed7.dev/p/building-agents-is-trivial-now-context-is-the-next-front-3c0b2ed056.md" } }, { "schema_version": "1.1", "id": "auto-83d31714cc", "slug": "unclecode-crawl4ai-83d31714cc", "url": "https://feed7.dev/p/unclecode-crawl4ai-83d31714cc", "title": "unclecode/crawl4ai", "why_included": "Update exposed Crawl4AI deployments: v0.9.2 fixes streaming leaks, while older releases had critical RCE, SSRF, and auth flaws.", "summary": "Crawl4AI turns browser-rendered pages into Markdown or structured data for agent and RAG pipelines. v0.9.2 fixes streaming leaks and deployment issues after two security-focused releases.", "practical_implication": "Builders using web retrieval should update rather than pinning older Docker images, keep API authentication enabled, and treat crawl requests as untrusted input. v0.9.0 made the Docker server secure by default, while crash recovery and prefetch support help longer jobs.", "agent_context": "Crawl4AI provides async browser crawling, Markdown generation, schema-based extraction, sessions, proxies, caching, deep crawls, and an MCP connection. **v0.9.2** fixes a streaming task/page leak plus Docker auth, packaging, and GPU-build issues.\n\nBuilders using web retrieval should update rather than pinning older Docker images, keep API authentication enabled, and treat crawl requests as untrusted input. **v0.9.0** made the Docker server secure by default, while crash recovery and prefetch support help longer jobs.\n\nThe broad feature set increases deployment and browser-state complexity. **v0.8.7** fixed critical RCE, SSRF, auth-bypass, file-write, XSS, and JWT-secret problems, so exposed legacy servers warrant particular scrutiny; the synchronous Selenium path is also deprecated.", "source": { "name": "GitHub", "url": "https://github.com/unclecode/crawl4ai", "published_at": "2026-08-30T00:00:00.000Z" }, "source_class": "tool", "content_type": "GitHub Repo", "layer": "tools", "domains": [ "coding", "data" ], "topics": [ "retrieval", "mcp", "tool-use" ], "verification": { "status": "needs_review", "label": "Needs Review", "method": "unverified", "verified_at": null }, "uncertainty": [ "Automatically selected from source material; feed7 has not independently tested the claim." ], "connected_context": null, "lifecycle": "New", "published_at": "2026-08-30T00:00:00.000Z", "modified_at": "2026-08-30T00:00:00.000Z", "supersedes": [], "expires_at": null, "formats": { "html": "https://feed7.dev/p/unclecode-crawl4ai-83d31714cc", "json": "https://feed7.dev/p/unclecode-crawl4ai-83d31714cc.json", "markdown": "https://feed7.dev/p/unclecode-crawl4ai-83d31714cc.md" } } ], "conflicting_sources": [], "superseded_claims": [], "corpus_evidence": [ { "title": "Large Language Models (LLMs) for Telecom Root Cause Analysis (RCA): A Structured Reasoning Framework for Evidence-Grounded Diagnosis", "url": "https://arxiv.org/abs/2609.02805v1", "source_name": "arXiv", "published_at": "2026-09-02T16:43:22+00:00", "summary": "A telecom diagnosis pipeline structures telemetry before forcing a decision path and evidence-grounded explanation. The pattern is relevant to agents reasoning over heterogeneous operational data." }, { "title": "DIASENTINEL: An Auditable Multi-Agent System for Guideline-Grounded Diabetes Risk Screening", "url": "https://arxiv.org/abs/2608.31128v1", "source_name": "arXiv", "published_at": "2026-08-31T17:40:43+00:00", "summary": "DIASENTINEL combines deterministic extraction, guideline retrieval, risk prediction, and hybrid verification on-premise. It is a useful architecture reference for auditable agents handling sensitive data." }, { "title": "Guardians of the State: An Air-Gapped AI Fortress for Consumer Data — Rachna Srivastava, DFPI", "url": "https://www.youtube.com/watch?v=2WZsT-znFTQ", "source_name": "AI Engineer", "published_at": "2026-08-29T15:00:25+00:00", "summary": "California’s financial regulator built an offline AI pipeline around replayable data, hardware-backed redaction, model routing, and one-way updates for court-defensible evidence." }, { "title": "How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked", "url": "https://www.youtube.com/watch?v=qdAkxLoYNI8", "source_name": "AI Engineer", "published_at": "2026-08-27T18:00:23+00:00", "summary": "Agents often stop after finding plausible code context. Werry argues for task-specific retrieval that also surfaces decisions, conventions, PR history, and expertise before planning or review." }, { "title": "How We Got LLMs to Recommend Our Open Source Library — Christopher Burns, Inth", "url": "https://www.youtube.com/watch?v=V_5bn4q-vAI", "source_name": "AI Engineer", "published_at": "2026-08-26T15:30:07+00:00", "summary": "For fast-changing libraries, ship concise agent-facing docs inside the package as well as on the web; agents may inspect installed code before visiting your site." }, { "title": "Inside 847 Production Clinical AI Notes — Sebastian Fox, Composo", "url": "https://www.youtube.com/watch?v=yqF6XhzbWBk", "source_name": "AI Engineer", "published_at": "2026-08-22T17:00:32+00:00", "summary": "Plausible outputs can hide consequential omissions that generic LLM judges miss. Production evals need real failure discovery and retrieved expert judgments, not a frozen rubric alone." }, { "title": "Re$^3$Cap: Retrieval-Guided Refinement for Image Captioning Enhancement via Reinforcement Learning", "url": "https://arxiv.org/abs/2608.21305v1", "source_name": "arXiv", "published_at": "2026-08-21T17:07:41+00:00", "summary": "Re³Cap uses multimodal retrieval to find caption omissions and hallucinations before refinement, offering a concrete retrieval-and-review pattern for vision agents." }, { "title": "Inject, Align, Recover: Staged Post-Training for Retrieval-Free Document Knowledge Internalization", "url": "https://arxiv.org/abs/2608.20281v1", "source_name": "arXiv", "published_at": "2026-08-20T17:14:24+00:00", "summary": "IAR turns a fixed document corpus into parametric knowledge through separate injection, QA alignment, and recovery stages. It improved domain QA while retaining more general capability than vanilla SFT." }, { "title": "Which Eviction Policy Should an LLM Cache Use? A Systematic Study Across Workloads, Capacities, and Encoders", "url": "https://arxiv.org/abs/2608.20280v1", "source_name": "arXiv", "published_at": "2026-08-20T17:14:19+00:00", "summary": "LFU was the strongest simple semantic-cache default, but answer validity mattered far more than eviction policy. Raw hit rates above 50% fell near 2% after quality adjustment." }, { "title": "Context Engineering in 2026 — Louis-François Bouchard, Omar Solano & Samridhi Vaid, Towards AI", "url": "https://www.youtube.com/watch?v=WP3hjUXd918", "source_name": "AI Engineer", "published_at": "2026-08-17T16:26:35+00:00", "summary": "In this AI-tutor evaluation, keeping full history beat compaction on recall and cost because cached tokens were cheap. Measure your constraint before shortening agent context." }, { "title": "How Web Data Infrastructure Powers the Next Generation of AI — Patricija Žemaitytė, Oxylabs", "url": "https://www.youtube.com/watch?v=1UmZHb_E_SM", "source_name": "AI Engineer", "published_at": "2026-08-14T17:00:37+00:00", "summary": "Fresh web retrieval becomes product infrastructure once agents need sub-second responses and sustained scale. The talk’s failed client demo shows why dev latency alone says little about production resilience." }, { "title": "SheetCompass: Hierarchical Relation Graphs for Agentic Spreadsheet Reasoning", "url": "https://arxiv.org/abs/2608.14452v1", "source_name": "arXiv", "published_at": "2026-08-14T16:39:16+00:00", "summary": "SheetCompass keeps spreadsheet structure as a hierarchical relation graph instead of flattening cells into text. Its agent memory retains task-relevant context across sheets, tables, and columns." }, { "title": "The Rise of CaaS: Context-as-a-Service for Agentic AI — Omer Primor, Bright Data", "url": "https://www.youtube.com/watch?v=Ot4OPrPH4xY", "source_name": "AI Engineer", "published_at": "2026-08-14T16:30:36+00:00", "summary": "A small company-enrichment test suggests rented web context is convenient for changing queries, while repeated stable queries may justify owning the pipeline. The reported crossover was around 15,000 queries." }, { "title": "Scaling Compute on Context — Jack Morris, Engram", "url": "https://www.youtube.com/watch?v=WiqDvX6isc4", "source_name": "AI Engineer", "published_at": "2026-08-12T15:30:14+00:00", "summary": "Training directly on a private corpus can memorize it without producing useful behavior. Durable context learning likely needs synthetic study tasks and recursive self-improvement, not a single fine-tuning pass." }, { "title": "Exa web search free through August 31 on AI Gateway and eve", "url": "https://vercel.com/changelog/exa-web-search-free-through-august-31-on-ai-gateway-and-eve", "source_name": "Vercel", "published_at": "2026-08-12T00:01:00+00:00", "summary": "AI Gateway now exposes Exa search as a keyless tool for any gateway model, and eve uses it by default. The free period ends August 31, so keep cost assumptions temporary." }, { "title": "PsychoAgent: An Affect-Sensitive Cognitive Architecture for Conflict-Aware Memory in LLM Agents", "url": "https://arxiv.org/abs/2608.07438v1", "source_name": "arXiv", "published_at": "2026-08-07T17:22:29+00:00", "summary": "PsychoAgent separates factual and affective memory, then reranks relevant memories by salience. It retrieved more conflict-critical context, but output-quality differences were not significant." }, { "title": "Teaching Nemotron Greek: Mining a Corpus, Adapting Retrieval, and Grounding Generation for Modern Greek across Specialist Domains", "url": "https://arxiv.org/abs/2608.05138v1", "source_name": "arXiv", "published_at": "2026-08-05T17:56:40+00:00", "summary": "A Greek RAG study shows generic multilingual embeddings can lose to BM25 in specialist domains, while targeted adaptation sharply improves retrieval and grounded answers. Language-specific evaluation still matters." }, { "title": "Structured Memory for Edge Language Models: Persistent Context and Corpus Retrieval via O(1) SSM State Injection", "url": "https://arxiv.org/abs/2608.02560v1", "source_name": "arXiv", "published_at": "2026-08-03T17:43:36+00:00", "summary": "PRECOG precomputes SSM corpus states and injects them at query time, avoiding context re-ingestion. On a 1.2B edge model, reported prefill fell from about 27 seconds to under 6 ms." }, { "title": "ReToken: One Token to Improve Vision-Language Models for Visual Retrieval", "url": "https://arxiv.org/abs/2607.28627v1", "source_name": "arXiv", "published_at": "2026-07-30T17:59:56+00:00", "summary": "ReToken adds one learned retrieval embedding to select relevant visual tokens from a cached context, improving long-image and video benchmarks while fitting training and inference on one H100." }, { "title": "Build for the Memo, Not the Demo — Shawn Chan, China Resources Holdings", "url": "https://www.youtube.com/watch?v=tJFjeMBKbIY", "source_name": "AI Engineer", "published_at": "2026-07-30T02:00:06+00:00", "summary": "Finance agents need claim-level provenance, explicit uncertainty, consistency checks, surfaced contradictions, and logged approval. Fluent output without those controls will not survive diligence." }, { "title": "UniMem: Complementary Episodic-to-Parametric Memory for Boundary-Agnostic Task Streams", "url": "https://arxiv.org/abs/2607.26017v1", "source_name": "arXiv", "published_at": "2026-07-28T17:28:21+00:00", "summary": "UniMem routes novel experience into retrieval memory and consolidates recurring patterns into expandable parameters, offering a design for agents that learn across unlabelled task streams." }, { "title": "MemLens: A Value-Aware Memory Management System with Interactive Analytics for LLM-based Agents", "url": "https://arxiv.org/abs/2607.25992v1", "source_name": "arXiv", "published_at": "2026-07-28T17:08:56+00:00", "summary": "MemLens scores individual agent memories, stores them by value, and exposes quality, latency, and token tradeoffs—an inspectable alternative to retaining every interaction equally." }, { "title": "Serving 2 Million Models Without Melting: Scaling the Hugging Face Hub — Arek Borucki, Hugging Face", "url": "https://www.youtube.com/watch?v=lyL5QhgIOxc", "source_name": "AI Engineer", "published_at": "2026-07-28T13:41:11+00:00", "summary": "Hugging Face’s Hub shows how to scale model search: precompute tokens, separate metadata from artifacts, isolate heavy reads, and autoscale from workload signals." }, { "title": "AI on Your Lakehouse: Context Comes in Shapes, Not Queries — Zach Blumenfeld, Neo4j", "url": "https://www.youtube.com/watch?v=kRkcNOsRyYg", "source_name": "AI Engineer", "published_at": "2026-07-23T07:00:23+00:00", "summary": "Graph-shaped context can expose relationships and document structure that vector search or Text-to-SQL misses. Treat it as an additional retrieval surface, then benchmark it against your own data." }, { "title": "Citation Needed: Provenance for LLM-Built Knowledge Graphs — Daniel Chalef, Zep AI", "url": "https://www.youtube.com/watch?v=H7puB0RwJMM", "source_name": "AI Engineer", "published_at": "2026-07-23T04:00:02+00:00", "summary": "LLM synthesis breaks simple source pointers when facts merge, change, or outlive one input. Model provenance inside the context graph so agents can filter trust, debug derivation, and apply deletion policies." }, { "title": "Graphify-Labs/graphify", "url": "https://github.com/Graphify-Labs/graphify", "source_name": null, "published_at": null, "summary": "Graphify gives coding agents a queryable project graph with provenance-tagged relationships, reducing repeated repository scans while keeping inferred links visibly distinct from extracted facts." }, { "title": "superlinked/sie", "url": "https://github.com/superlinked/sie", "source_name": "GitHub", "published_at": null, "summary": "SIE consolidates retrieval, document conversion, structured output, safety, and generation models behind one self-hosted, OpenAI-compatible API for agent stacks." }, { "title": "Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation", "url": "https://arxiv.org/abs/2607.05382v1", "source_name": null, "published_at": null, "summary": "SearchGen-Bench shows open image generators score 21–28/100 on long-tail entities, and naive search retrieval only adds noise; a teach-then-search co-training recipe learns when to retrieve versus rely on weights." }, { "title": "virgiliojr94/book-to-skill", "url": "https://github.com/virgiliojr94/book-to-skill", "source_name": "GitHub", "published_at": null, "summary": "book-to-skill compiles books and document sets into on-demand agent skills, reducing repeated context loading while preserving chapter-level references and reusable decision rules." }, { "title": "AgriciDaniel/claude-obsidian", "url": "https://github.com/AgriciDaniel/claude-obsidian", "source_name": "GitHub", "published_at": null, "summary": "A local-first Obsidian system gives coding agents durable, cited context while keeping Markdown portable and serializing writes into recoverable transactions." }, { "title": "thedotmack/claude-mem", "url": "https://github.com/thedotmack/claude-mem", "source_name": "GitHub", "published_at": null, "summary": "Claude-Mem records agent activity, compresses it into persistent project memory, and retrieves selected history later, trading extra local services and stored data for cross-session continuity." }, { "title": "ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning", "url": "https://arxiv.org/abs/2607.02509v1", "source_name": null, "published_at": null, "summary": "ReContext is a training-free harness that replays query-relevant evidence from long inputs before answering, taking the best average rank across 8 long-context benchmarks up to 128K on Qwen3-4B/8B and Llama3-8B." }, { "title": "infiniflow/ragflow", "url": "https://github.com/infiniflow/ragflow", "source_name": "GitHub", "published_at": null, "summary": "RAGFlow packages document ingestion, retrieval, citations, agent workflows, memory and MCP into one self-hostable context layer. It is useful when an agent needs traceable answers across mixed private data." }, { "title": "Panniantong/Agent-Reach", "url": "https://github.com/Panniantong/Agent-Reach", "source_name": "GitHub", "published_at": null, "summary": "Agent Reach installs, checks, and routes web-access tools for coding agents, reducing the setup work behind research across social platforms, video sites, GitHub, RSS, and the open web." }, { "title": "tirth8205/code-review-graph", "url": "https://github.com/tirth8205/code-review-graph", "source_name": "GitHub", "published_at": null, "summary": "code-review-graph gives coding agents a local, incrementally updated dependency map over MCP, narrowing review context while exposing affected callers, dependents, and tests." }, { "title": "How Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation?", "url": "https://arxiv.org/abs/2607.11783v1", "source_name": null, "published_at": null, "summary": "A RAG study finds that retrieved ideology carries into answers and varies with sampling temperature, so source audits and decoding tests should be evaluated together." }, { "title": "RLM: Recursive Language Models for Large Codebases - Shashi, Superagentic AI", "url": "https://www.youtube.com/watch?v=8oyalrfwgjw", "source_name": null, "published_at": null, "summary": "RLMs treat a large repository as external data that an agent inspects with code, returning bounded evidence to the main context instead of loading or summarizing everything upfront." } ] }