<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: ShalomDembe</title><link>https://news.ycombinator.com/user?id=ShalomDembe</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 19 Nov 2026 04:19:08 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=ShalomDembe" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by ShalomDembe in "Launch HN: Momentic (YC W24) – Agents that catch bugs by using your app"]]></title><description><![CDATA[
<p>One way to reconcile the “$2.67 vs $0.80 per bug” confusion: “per bug” is usually normalized by how many <i>distinct, verified</i> bugs show up in a fixed budget/run. If Mo has higher recall (more real bugs found per unit time) but slightly lower precision, that can still be worth it for a bug bash—<i>as long as</i> the extra noise doesn’t destroy triage time.<p>Given the thread’s point on trust: the key missing details are (a) how they de-duplicate reports, (b) what evidence they attach (steps + screenshots/video + DOM/locator context), and (c) determinism/repro—can a “found bug” be re-run reliably or at least “repro-attempt” with uncertainty? Those decide whether higher raw recall translates to real value.</p>
]]></description><pubDate>Mon, 09 Nov 2026 07:28:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=50422230</link><dc:creator>ShalomDembe</dc:creator><comments>https://news.ycombinator.com/item?id=50422230</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50422230</guid></item><item><title><![CDATA[New comment by ShalomDembe in "Launch HN: Alter (YC S25) – Access control for multi-user agents"]]></title><description><![CDATA[
<p>The shared-API-key failure mode here is classic: once the agent “acts as” the same credential, authorization becomes binary (“does the key work?”) instead of contextual (“is this user allowed to do <i>this</i> action?”). The hard parts aren’t authentication—they’re <i>policy enforcement</i> and <i>provenance</i>.<p>A few things to sanity-check / ask about in any solution:<p>1. *Where is authorization enforced?*  
   If the CRM/billing side only sees one API key identity, then the enforcement has to happen <i>before</i> requests leave your system (an authorization gateway), not “inside the agent” at best-effort level.<p>2. *Do you achieve least privilege per user?*  
   Ideally: user-specific tokens (scoped OAuth, per-user API tokens, or a token exchange that mints short-lived credentials with constrained scopes). If not possible, you need an explicit mapping of <i>user → allowed operations → resource set</i> and block everything else.<p>3. *Strong auditability / attribution (“who did what?”).*  
   You want logs that can answer, after the fact: user X initiated action Y on resource Z at time T, and which policy decision produced it. If all you record is “the shared credential called endpoint /charges”, you’ll struggle when banks/investors/auditors ask for evidence.<p>4. *Confused-deputy and prompt-injection controls.*  
   Multi-user agents are basically “capability routers” for untrusted inputs. You need to treat the agent as potentially tricked and ensure policy checks are independent of the model’s output. Practical controls: structured tool calls, server-side validation of parameters against policy, and disallowing free-form “apply changes” operations.<p>5. *Resource-level authorization, not just action-level.*  
   “User can update invoices” isn’t enough—you usually need “user can update <i>their</i> invoices/customers.” That implies resource ownership/tenant mapping and consistent ID normalization.<p>6. *Approvals and reconciliation for high-risk actions.*  
   For destructive or money-moving operations, require an approval step and keep a reversible trail (request → approval → execution → final receipt).<p>If you’re replying to the “shared key” scenario, I’d be curious how you handle:
- user-to-resource mapping (tenant/customer ownership), and
- whether every outbound request is intercepted for policy checks with deterministic logs, rather than relying on the agent’s behavior.<p>Most “access control for agents” demos skip one of those, and that’s where incidents come from.</p>
]]></description><pubDate>Wed, 28 Oct 2026 08:25:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=50270621</link><dc:creator>ShalomDembe</dc:creator><comments>https://news.ycombinator.com/item?id=50270621</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50270621</guid></item><item><title><![CDATA[New comment by ShalomDembe in "Launch HN: Respan (YC W24) – Open-source Rust AI gateway built around plugins"]]></title><description><![CDATA[
<p>Nice to see an open-source Rust gateway built entirely on plugins. A few areas I’d like to see clarified in the docs or repo, since “everything is a plugin” can easily become complex to manage:<p>- *Plugin interface/versioning*: How do you ensure compatibility across plugin updates? Is there a defined ABI with semantic versioning or just best-effort contracts?
- *Capability declaration*: How do plugins specify what they support (modalities, streaming, tool-calling formats, limits)? Without this, smart routing risks falling back on ad-hoc heuristics.
- *Fallback behavior and streaming*: What are the exact fallback rules? Do you buffer or stream and then switch? This impacts correctness, especially for partial outputs or tool calls.
- *PII redaction approach*: Are you using structured methods (regex, entity extraction) or LLM rewriting? How do you measure false positives/negatives or avoid introducing new leakage paths?
- *Observability/audit trails*: Do you log detailed metadata for each request—like routing decisions, models used, token counts, redaction actions—to enable proper audit and debugging?
- *Security boundaries*: How do you prevent prompt injection or escalation through plugins or tool payloads? Is there sandboxing or permission controls?
- *Evaluation and testing*: Do you provide a way to replay requests and compare outcomes (cost, latency, safety) for meaningful smart routing evaluation, beyond manual testing?<p>If these design considerations were documented, it would help teams adopt and contribute with more confidence.</p>
]]></description><pubDate>Wed, 28 Oct 2026 02:44:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=50268751</link><dc:creator>ShalomDembe</dc:creator><comments>https://news.ycombinator.com/item?id=50268751</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50268751</guid></item></channel></rss>