Hacker News: niyikiza

New comment by niyikiza in "Ask HN: AI Agent and harness containerization/security recommendations"

niyikiza — Thu, 16 Jul 2026 16:58:58 +0000

We built an open source governance tool called Tenuo that allows you to define exactly what an agent can do for a given tasks. Works for local filesystem, network, mcp servers, etc.

New comment by niyikiza in "Ask HN: What Are You Working On? (July 2026)"

niyikiza — Mon, 13 Jul 2026 07:37:52 +0000

I'm working on a Claude Code governance tool that allows to define deterministic policies for tool call that can be enforced across a fleet and will be in effect even when individual users run with --dangerously-skip-permissions

https://github.com/tenuo-ai/claude-governance

New comment by niyikiza in "When AI Costs More Than the Engineer"

niyikiza — Mon, 06 Jul 2026 08:36:05 +0000

I guess they should include tuition cost as well.

New comment by niyikiza in "A way to exclude sensitive files issue still open for OpenAI Codex"

niyikiza — Sun, 28 Jun 2026 22:45:56 +0000

We've been using Tenuo which for task-scoped authorization.

Its integration for Claude Code: https://github.com/tenuo-ai/claude-governance

New comment by niyikiza in "LastPass notifies users of yet another data breach"

niyikiza — Thu, 25 Jun 2026 16:04:32 +0000

Because procurement is hard. Changing vendors is a big undertaking for big companies. They are certainly not going to be switching vendors every time there is an incident

New comment by niyikiza in "OAuth for all"

niyikiza — Thu, 25 Jun 2026 08:49:38 +0000

Means you can basically host your own AS

New comment by niyikiza in "OAuth for all"

niyikiza — Thu, 25 Jun 2026 08:48:32 +0000

there are some emerging mechanisms for offline verification that don't require AS in the OAuth WG. (I'm working on one of them)

New comment by niyikiza in "Building reliable agentic AI systems"

niyikiza — Sun, 21 Jun 2026 06:55:19 +0000

What would the benefit be? A mega agent that does everything?

There are some well documented advantages of decomposition...that's why the industry favours microservices over monoloths.

New comment by niyikiza in "Zero-Touch OAuth for MCP"

niyikiza — Fri, 19 Jun 2026 03:24:32 +0000

I agree with the coarse permissions point, and I wouldn't bank on those services adding finer-grained scopes.

The idea in my draft is to do the attenuation and verification before the call reaches the service, enforced at the boundary, like the proxy setup you're describing. And the token wouldn't be a bearer token per se; there's proof of possession, and the constraints narrow at each hop and travel with the token, so the boundary can verify the chain itself rather than rely on a central authority. The design is inspired by macaroons and other capability-based access control work.

Full draft's here if you want to pick it apart: https://datatracker.ietf.org/doc/draft-niyikiza-oauth-attenu...

New comment by niyikiza in "Zero-Touch OAuth for MCP"

niyikiza — Fri, 19 Jun 2026 02:35:11 +0000

There's some active discussions on task level authz and multi-hop delegation in the OAuth WG right now. WorkOS wrote a good overview of the open drafts [1]. (Disclosure: one of them is mine.) [1] https://workos.com/blog/oauth-multi-hop-delegation-ai-agents

Zero-Touch OAuth for MCP

niyikiza — Thu, 18 Jun 2026 21:54:10 +0000

Article URL: https://blog.modelcontextprotocol.io/posts/enterprise-managed-auth/

Comments URL: https://news.ycombinator.com/item?id=48592163

Points: 278

# Comments: 103

New comment by niyikiza in "Hermes Agent is now natively supported on Windows"

niyikiza — Mon, 01 Jun 2026 04:56:59 +0000

Some clever workarounds wrt process management & POSIX compatibility

New comment by niyikiza in "Why Japanese companies do so many different things"

niyikiza — Fri, 22 May 2026 18:01:02 +0000

Agree with the meta point. I worked in Korea and Japan and loved the culture but when I moved to the west I was surprised to see how people over here fantasize about their (imo inefficient) corporate cultures.

This particular article was decently nuanced though.

New comment by niyikiza in "GitHub is investigating unauthorized access to their internal repositories"

niyikiza — Wed, 20 May 2026 03:40:19 +0000

My understanding is that when it's something that requires user action they'd directly send comms to customers.

New comment by niyikiza in "GitHub is investigating unauthorized access to their internal repositories"

niyikiza — Wed, 20 May 2026 03:39:16 +0000

Probably the best option after sending a mass email when customers need to take action. The status page is for reliability issues impacting end users & the blog is for in-depth analysis.

New comment by niyikiza in "Ask HN: What are you working on? (May 2026)"

niyikiza — Sun, 10 May 2026 22:19:41 +0000

Building tenuo.ai (https://github.com/tenuo-ai/tenuo): task-scoped authorization for AI agents. Rust implementation of capabilities + cryptographic offline verification.

New comment by niyikiza in "Agents need control flow, not more prompts"

niyikiza — Thu, 07 May 2026 23:19:29 +0000

My analogy[1] has been that we need a valet key: capped speed, geofenced, short ttl, can't open trunk/glovebox, etc. That way we don't have to say pretty please to the valet and hope that they won't get ideas.

[1] https://niyikiza.com/posts/capability-delegation/

New comment by niyikiza in "An AI agent deleted our production database. The agent's confession is below"

niyikiza — Mon, 27 Apr 2026 04:21:40 +0000

I have to agree here...of all things that went wrong here, I don't think the API surface is to blame. You need to have deterministic control & escalation mechanism on your agents whether they are calling an API or any other tool

New comment by niyikiza in "US special forces soldier arrested after allegedly winning $400k on Maduro raid"

niyikiza — Sat, 25 Apr 2026 04:00:54 +0000

I SAY AYE.

New comment by niyikiza in "Tell HN: Claude 4.7 is ignoring stop hooks"

niyikiza — Sat, 25 Apr 2026 03:57:56 +0000

Yeah, people calibrate trust to the median behaviour of the model and get burned by the tail. What makes it harder is that even people who do see the holes often respond with better prompts and more elaborate context. Same trust-the-model move one level up. Hyperscalers aren't incentivized to fight that instinct either. Every "fix" routes more tokens through their meter.