Feed aggregator

Article URL: https://rijnard.com/blog/dear-agent-proof

Comments URL: https://news.ycombinator.com/item?id=46984891

Points: 1

# Comments: 0

Article URL: http://ternarysearch.blogspot.com/2026/02/reflections-on-using-claude-code.html

Comments URL: https://news.ycombinator.com/item?id=46984890

Points: 1

# Comments: 0

Article URL: https://blog.janestreet.com/advent-of-fpga-challenge-2025-results/

Comments URL: https://news.ycombinator.com/item?id=46984865

Points: 2

# Comments: 0

Article URL: https://www.helpnetsecurity.com/2023/07/05/mike-fey-island-enterprise-browser/

Comments URL: https://news.ycombinator.com/item?id=46984854

Points: 1

# Comments: 0

Article URL: https://thebrunelmuseum.com/engineering-connections/

Comments URL: https://news.ycombinator.com/item?id=46984828

Points: 2

# Comments: 1

Article URL: https://self-distillation.github.io/SDFT.html

Comments URL: https://news.ycombinator.com/item?id=46984810

Points: 1

# Comments: 0

Article URL: https://github.com/b4rtaz/distributed-llama

Comments URL: https://news.ycombinator.com/item?id=46984802

Points: 3

# Comments: 0

Article URL: https://glm5.net/

Comments URL: https://news.ycombinator.com/item?id=46984799

Points: 4

# Comments: 1

Hi HN,

I built Prompt Builder (https://www.promptbuilder.space/) because I was frustrated managing complex AI prompts in plain text boxes.

The core idea: instead of one giant textarea, you compose prompts from draggable, reorderable blocks. Each block can have its own XML/custom tags, visibility toggles (so you can A/B test sections), and duplicates.

Key technical details: - Live compiled preview: the right pane shows the exact string being sent, updating as you drag/toggle blocks - Dynamic variables: define {{var_name}} once, use across blocks, change in one place - Multi-model execution: send the same compiled prompt to different model APIs and compare outputs - Voice transcription + optional translation (useful for non-English prompt authoring) - Editable response pane: modify the model's output in place - Folder organization, search, and password-protected shareable links

I'd love feedback on the UX and whether the block metaphor resonates with how you think about prompt structure. What's missing?

Comments URL: https://news.ycombinator.com/item?id=46984792

Points: 1

# Comments: 0

Article URL: https://www.npr.org/2026/02/11/nx-s1-5552216/james-van-der-beek-dead-dawsons-creek

Comments URL: https://news.ycombinator.com/item?id=46984790

Points: 3

# Comments: 0

Article URL: https://github.com/policygate/clawshield

Comments URL: https://news.ycombinator.com/item?id=46984787

Points: 1

# Comments: 2

Article URL: https://oedmethod.substack.com/p/best-conversations-are-in-cars

Comments URL: https://news.ycombinator.com/item?id=46984786

Points: 6

# Comments: 0

Article URL: https://apps.apple.com/us/app/shelver-home-organization/id6756636954

Comments URL: https://news.ycombinator.com/item?id=46984781

Points: 2

# Comments: 2

Article URL: https://hackaday.com/2026/02/11/motorolas-password-pill-was-just-one-idea/

Comments URL: https://news.ycombinator.com/item?id=46984778

Points: 1

# Comments: 0

Article URL: https://www.heritagedaily.com/2026/02/1300-year-old-world-chronicle-unearthed-in-sinai/156948

Comments URL: https://news.ycombinator.com/item?id=46984768

Points: 2

# Comments: 0

Membrane is a general-purpose memory substrate for agentic systems.

Most agent memory today is either context window state or an append only retrieval store. That allows recall but not learning. Facts become stale, procedures drift, and agents cannot revise knowledge safely without losing provenance or auditability.

Membrane treats memory as something that can evolve over time. It ingests raw experience, promotes it into structured knowledge, and allows records to be superseded, forked, contested, merged, or retracted with evidence. The goal is to let long lived agents improve while remaining predictable and inspectable.

The core ideas are simple.

- Memory is typed instead of stored as flat text - Knowledge is revisable and keeps provenance - Agents can learn competences and procedures, not just facts - Salience decays over time so unused knowledge fades - Retrieval is filtered by trust and sensitivity levels - Revision history remains auditable

Membrane runs either as a daemon with a gRPC API or as an embedded Go library. Storage uses SQLite with optional SQLCipher encryption. The repository includes an evaluation suite covering ingestion, revision, consolidation, and retrieval ordering.

Membrane intentionally does not implement vector similarity search. Retrieval backends and agent policy are separated from the memory layer so the system can remain deterministic and inspectable.

I built this while experimenting with long lived agents that need something closer to learning than what RAG systems currently provide. Feedback on architecture, edge cases, and real-world use cases would be helpful.

If you have any suggestions or want to contribute, anything is welcome :)

Comments URL: https://news.ycombinator.com/item?id=46984655

Points: 1

# Comments: 0

Article URL: https://www.cnn.com/2026/02/10/tech/google-video-nancy-guthrie

Comments URL: https://news.ycombinator.com/item?id=46984652

Points: 1

# Comments: 1

Article URL: https://github.com/ToheedAsghar/DocForge

Comments URL: https://news.ycombinator.com/item?id=46984641

Points: 1

# Comments: 0

Hi HN, I built an assumption-driven AI security assessment for teams shipping AI features without a dedicated security team yet.

You paste your AI use case (what it does, data types, vendors, deployment). In ~10 minutes you get a PDF report by email containing: - Trust boundaries + data flows + a threat model diagram (explicitly marked as conceptual/assumption-based) - Threats mapped to STRIDE + MAESTRO (agentic AI) - A risk rating (impact/likelihood) + 5×5 risk matrix - Recommended security controls and compliance mappings (example: EU AI Act, NIST AI 600-1)

Important: we make assumptions (ex: AWS deployment, common patterns) and we call them out in the report so you can correct them.

Link: https://raxit.ai/assessment

Would love feedback on what’s wrong, what’s missing, and what would make this actually useful in a real security review.

Comments URL: https://news.ycombinator.com/item?id=46984636

Points: 1

# Comments: 0