Research
Open-source software and articles from Stack Research.
- Apr 13, 2026 research opinion Spectacle, Silence, Calcification: The Governance Problem Hiding Inside Every Technology Hype Cycle The public spectacle surrounding artificial intelligence is not where the consequential decisions are being made.
- Apr 10, 2026 security identity NHI and Agentic Risk: Third-Party Tools Every tool an agent invokes runs someone else's code with your credentials. That's the supply chain problem.
- Apr 5, 2026 research security A Field Guide to the Wilderness The art of surviving interactions with text, and other artifacts, from the outside world. Ordinary hygiene applied to a workflow that grew faster than its controls.
- Apr 2, 2026 oss engineering Structural Debugging for Chain-of-Thought Graphs Building a tool that treats chain-of-thought reasoning as a program to debug, not a process to steer. It finds where the logic broke.
- Mar 29, 2026 research security Agent Security Is a Release Engineering Problem Risk is often created between changes, not inside one change. Agent systems become dangerous when short-lived input hardens into durable memory and survives longer than the assumptions that made it safe.
- Mar 23, 2026 research engineering Why Agent Memory Needs a Control Plane Agent systems need a memory control plane because retrieval safety and historical truth require enforceable policy and clear lineage, not mutable notes.
- Mar 17, 2026 oss engineering Executable Metaphors: A Compiler Where the Source Code Is an Analogy We built a compiler that turns natural-language metaphors into running programs. The metaphor is the source of truth; the code is the build artifact.
- Mar 16, 2026 oss engineering The Unaskable Question We built a tool that asks LLMs structurally impossible questions and classifies how they fail. The failures are more interesting than we expected.
- Mar 15, 2026 oss engineering Evolving Better Prompts An open source tool that breeds, mutates, and selects LLM prompts across generations.
- Mar 14, 2026 oss engineering Memory Should Decay We gave an agent 50 memories and let them decay. After 30 ticks, it retained 8 — exactly the ones it was still using.
- Mar 14, 2026 engineering research ControlOps: Letting Machines Talk The previous Let Machines Talk posts described the theory. This is what it looks like as running code.
- Mar 13, 2026 engineering research Let Machines Talk: Rollback Stopping the system is half the problem. Undoing what it already did is the other half.
- Mar 12, 2026 engineering research Let Machines Talk: Decision Lineage Observability tells you what's happening. Lineage tells you why.
- Mar 11, 2026 engineering research Let Machines Talk: Containing Failure The question isn't whether your system will fail. It's how much it takes down when it does.
- Mar 10, 2026 engineering research Let Machines Talk: Kill Paths A kill switch is a button. A kill path is an engineering discipline.
- Mar 9, 2026 security opinion Agents Get Socially Engineered Too Prompt injection is one attack shape. Authority theater, urgency pressure, and trust laundering across tool boundaries are the rest.
- Mar 8, 2026 research engineering Build for the Hour After Failure Agent systems get designed for launch day. The first hour after failure is usually an afterthought.
- Mar 1, 2026 opinion research Let Machines Talk Machine-to-machine communication isn't the risk. The risk is building the channels without building the controls.
- Mar 1, 2026 research security Earn the Right to Touch Production An agent that can produce a good plan is not the same as an agent that should be allowed to run it.
- Feb 28, 2026 research engineering Stress-Test the Plan, Not Just the Model Before an agent acts, force its plan through a thousand bad futures. Ship what survives.
- Feb 27, 2026 research engineering Intelligence Beyond Autocomplete Five directions for AI systems that aren't built on token prediction: deterministic reasoning, reversible execution, constrained testing, native substrates, and symbolic computation.
- Feb 27, 2026 research engineering AI That Refuses to Predict What if an AI system never produced text at all — just causal graphs, state machines, and consistency proofs?
- Feb 24, 2026 security identity NHI and Agentic Risk: When Humans Use Machine Credentials When people use machine credentials, intent is blurred and audit trails break. Agents make it worse.
- Feb 21, 2026 security identity NHI and Agentic Risk: Blast Radius Engineering Reused identities and weak isolation turn local incidents into systemic ones.
- Feb 17, 2026 security identity NHI and Agentic Risk: Secrets, Memory, and Persistence Leaked secrets become durable agent memory. That's why breaches linger.
- Feb 16, 2026 oss engineering Software That Expires EntropyOS is a Python runtime where APIs, code paths, and state have expiration dates built in.
- Feb 15, 2026 security oss A Real ASI02 Gap We Caught Before Shipping An adversarial test exposed unsafe text propagation across agent boundaries. We found it, fixed it, and kept the test.
- Feb 14, 2026 security identity NHI and Agentic Risk: Least Privilege Meets Least Agency Over-scoped identities turn harmless tools into high-impact actions. The agent is rarely the root problem.
- Feb 13, 2026 oss education Learn Security by Playing It An open-source collection of text adventure games that teach security, ops, and incident response through interactive scenarios.
- Dec 26, 2025 security identity NHI and Agentic Risk: How Compromise Happens Agent incidents usually start with a mundane identity fault, not a novel model exploit.
- Dec 24, 2025 news guides AI News Sources for 2026 A short list of news sites, blogs, and newsletters worth following for AI developments in 2026.