All the research, one tab.
Every research item behind the daily briefs, newest first, in fast pages. Filter by topic, open any card's source, share any page — the URL is the state. The live map stays on the Brief page.
- Market & Business
Microsoft's 'agentic SOC' keeps the human — and changes the job to setting the thresholds
Even for a one- or two-person team the pattern holds: automate the mechanical, reserve human judgment for the irreversible, and set explicit thresholds for what an agent may do unattended.
- AI in Education✓ verified
Khanmigo scale + Sal Khan's candid "non-event" reflection
The most-watched competitor's honest signal that scale does not equal impact — structure and motivation design matter more than model access. Exactly the gap AI Uni's structured-lesson model targets.
- Dev Tooling & Infra✓ verified
GitHub lets you assign a dependency alert straight to an AI agent to fix
Dependency triage is fatigue-heavy toil an agent can genuinely take off your plate — as long as the merge stays a human decision.
- Dev Tooling & Infra✓ verified
A poisoned npm package quietly rewrote a coding agent's memory — and it reloaded every session
Treat any automatic edit to an agent's memory or instruction files as a reviewable event, not a silent auto-load — a single poisoned dependency can otherwise steer every future run.
- Research✓ verified
A survey and paper list mapping agent-memory architectures, so you don't have to design one blind
Agent memory is the least-settled piece of most production agent stacks — a maintained survey and paper list is the fastest way to see which memory architecture actually matches your agent's failure mode before building a bespoke one from scratch.
- AI in Education✓ verified
Khan + TED + ETS launch AI-focused college (Khan TED Institute)
A direct competitive signal for AI Uni ("the college alternative for the AI economy") — a credentialed-degree entrant in applied-AI education.
- Dev Tooling & Infra✓ verified
mem0 / agent-memory architectures maturing
AI Uni's own memory-architecture work sits in this fast-moving category — the anti-evaporation problem the platform substrate solves.
- Dev Tooling & Infra
KV-cache agent-state persistence: reported 89% better completion, 67% fewer calls
If the effect holds, strong evidence for on-disk substrate-loading + persistence work. UNVERIFIED effect size — find the primary benchmark before citing the numbers.
- Dev Tooling & Infra✓ verified
Agent observability field: LangSmith vs Braintrust vs Langfuse vs Arize
Informs deterministic-vs-LLM-judge layering. Braintrust's merge-blocking eval-action is a pattern to study — but LLM-judge stays post-hoc, never replacing deterministic CI gates.
- MCP & Interop✓ verified
MCP's own 2026 roadmap shows where the protocol still has rough edges
MCP's own roadmap is the clearest signal of where the interoperability protocol still has rough edges — teams evaluating it for production should track SSO and audit-trail readiness before betting deployment plans on it.
- MCP & Interop✓ verified
24,008 secrets found in public MCP config files — 2,117 of them still live
The convenience of pasting a key into an MCP config is exactly how it leaks — keep secrets out of tool configs, and read a tool's description as untrusted input, not just documentation.
- Research✓ verified
Agentic Context Engineering (evolving contexts for self-improving LMs)
The academic framing of what the substrate-batch + landscape-brief + KB-principle loop does informally. Indexed via a curated list — read the underlying papers before teaching specifics.
- Agent Frameworks & Orchestration✓ verified
Four ways to orchestrate agents just quietly became the standard menu
Four orchestration styles stabilizing means picking a framework is now a real architecture decision, not a bet on which project survives — match your agent's actual coordination problem (graph, role, handoff, or hierarchical) before committing to one.
- Dev Tooling & Infra✓ verified
'ClawHavoc': 900+ malicious agent 'skills' pushed through a third-party marketplace
Any skill, tool, or config you didn't write deserves the same scrutiny as a code dependency — and 'delete the skill' is not the same as 'undo what it did.'
- MCP & Interop✓ verified
MCP now spoken natively by every major host; 500+ public servers
Confirms MCP as the durable interop bet for Intel's agent-facing side — "serve our KB over MCP" reaches every major client without per-client integration.
- Agent Frameworks & Orchestration✓ verified
OWASP's agentic security Top 10 turns 'excessive agency' into a least-agency discipline
If your product lets an agent take actions, this is the checklist to threat-model against: give it the least autonomy the job needs, gate the state-changing steps behind a human, and cap how long it can loop.
- AI in Education✓ verified
UK LearnLM classroom RCT — independent corroboration
A second independent RCT with transfer-to-novel-problems as the outcome — the hard test — strengthening the evidence base beyond a single study or vendor.
- AI in Education✓ verified
AI tutoring keeps beating active learning in RCTs
The empirical floor under AI Uni's whole pedagogical thesis — structured AI tutoring over traditional pedagogy. Verify specific effect-size figures against the primary before re-quoting.