frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN:AIP Protocol–Solving the agent revocation problem in distributed systems

https://github.com/theaniketgiri/aip
1•theaniketgiri•1h ago
I realized something uncomfortable while running agents in production:

APIs authenticate the process making a request.

But with LLM agents, the process no longer decides the request — the model does.

So when an agent is prompt-injected or misaligned, authentication still succeeds. The system verifies who executed the call, not who chose it.

Rotating keys or adding revocation checks doesn’t fix this. You’re still trusting the wrapper while the decision lives inside the model.

The missing primitive isn’t stronger identity — it’s verifying the action itself.

In a typical backend system:

service → calls API auth verifies which process made the call

This works because the process contains the decision logic.

With agents:

model decides the action → process just executes it

So authentication still proves who called but no longer proves who decided

If an agent is compromised (prompt injection, tool misuse, leaked context), rotating API keys or checking revocation lists doesn’t actually solve the problem — the system still trusts the process identity while the decision authority lives inside the model.

What we needed was verification of the action itself.

I built a small protocol where every side-effect requires a signed “intent”.

Each agent has a keypair. Every tool call carries a signature over:

action

parameters

timestamp

nonce (replay protection)

declared capabilities

Verification is local (Ed25519, <1ms). Services don’t call an auth server — they verify the intent.

Revocation is async: services subscribe to key invalidation events, but verification still works offline.

Example:

from aip_protocol import shield

@shield(actions=["read_db","send_email"]) class SupportAgent: def handle_ticket(self, ticket_id): ...

If the agent attempts an undeclared action, the call is rejected before execution.

Tradeoffs:

more complex than API keys (key management)

doesn’t stop prompt injection, only limits consequences

requires thinking in capabilities instead of identities

I’m trying to figure out if this is actually a missing primitive for agent systems or unnecessary complexity.

Repo: https://github.com/theaniketgiri/aip

Spec: https://github.com/theaniketgiri/aip/blob/master/RFC-001.md

Would especially like feedback from people running agents beyond demos.

Building Custom Docker Sandboxes

https://substack.com/home/post/p-188153139
1•shelajev•45s ago•0 comments

Bengt Hires a Human–Towards a Happy Future with AI Employers

https://andonlabs.com/blog/bengt-hires-a-human
1•lukaspetersson•1m ago•1 comments

Russian state media meddles in Swiss public broadcasting referendum

https://www.20min.ch/story/halbierungsinitiative-russisches-staatsmedium-mischt-sich-in-srg-absti...
1•leohoferdev•2m ago•0 comments

Deploy your OpenClaw agent in 5 minutes

https://fastclaw.ai/
1•idoubi•2m ago•0 comments

I Joined the MariaDB Foundation

https://lefred.be/content/i-joined-the-mariadb-foundation/
1•eatonphil•3m ago•0 comments

A Love Letter to Self-Hosting

https://lukaswerner.com/post/2026-02-13@self-hosting-letter
1•chilipepperhott•3m ago•0 comments

If AI writes most of the code, understanding codebases becomes the bottleneck

https://app.tryarchaic.com/
1•baijan•3m ago•1 comments

Break Stasis

https://oldmanrahul.com/2026/02/15/break-stasis/
1•oldmanrahul•3m ago•0 comments

Undetected Past Contacts with Technological Species and Technosignature Science

https://iopscience.iop.org/article/10.3847/1538-3881/ae394b
1•bikenaga•3m ago•0 comments

Password managers less secure than promised

https://ethz.ch/en/news-and-events/eth-news/news/2026/02/password-managers-less-secure-than-promi...
5•winterdeaf•3m ago•0 comments

Trying New Things

https://daoudclarke.net/2026/02/16/trying-new-things
2•daoudc•4m ago•0 comments

macOS Tahoe Finder Bug Underscores Apple's Slipping UI Polish

https://www.macrumors.com/2026/02/13/macos-tahoe-finder-bug-slipping-ui-polish/
2•akyuu•6m ago•0 comments

Google warns EU against 'erecting walls' in tech sovereignty push

https://www.ft.com/content/0847914c-be27-4573-8600-8cdb54e604b7
2•spiffyk•6m ago•1 comments

How to take a photo with scotch tape (lensless imaging) [video]

https://www.youtube.com/watch?v=97f0nfU5Px0
3•surprisetalk•7m ago•0 comments

GrowthClaw: Marketing workflows for OpenClaw with evaluation gates

https://github.com/mrrkrieg/growthos
3•dankrieg•8m ago•2 comments

Unitree's humanoid robot team's performance at the 2026 Spring Festival Gala

https://twitter.com/cyberrobooo/status/2023378370592174272
3•DustinEchoes•8m ago•0 comments

Programming a 144-computer chip to minimize power (2013) [video]

https://www.youtube.com/watch?v=0PclgBd6_Zs
2•tosh•9m ago•0 comments

Show HN: CabbageSEO: Check if AI mentions your business, then fix it if not

https://www.cabbageseo.com/
2•arjun060601•9m ago•0 comments

Show HN: Comfy Pilot – MCP server that lets Claude Code edit ComfyUI workflows

https://github.com/ConstantineB6/comfy-pilot
2•0xConstantine•10m ago•0 comments

(Un)portable defer in C

https://antonz.org/defer-in-c/
1•birdculture•12m ago•0 comments

Dyslexia, Programming and Lisp

https://www.iwillig.me/blog/on-dyslexia-and-lisp/
2•_emacsomancer_•14m ago•0 comments

Integration patterns: How we connect software

https://staffbase.com/blog/integration-patterns
2•goblin89•15m ago•0 comments

Architecting AI-ready infrastructure for the agentic era

https://thenewstack.io/ai-ready-infrastructure/
1•dmk•16m ago•0 comments

What's Your Attention Worth? – The Ad Spend Calculator

https://attentionworth.com/
1•thunderbong•16m ago•0 comments

A Historical Reference of React Criticism

https://www.zachleat.com/web/react-criticism/
1•ishandotpage•16m ago•0 comments

Show HN: Hackable Skinny Clawdbot for Telegram

https://github.com/vseplet/smith
1•vseplet•16m ago•0 comments

Show HN: An beautiful webpage I made

https://github.com/adityaprasad-sudo/ExploreSingapore
1•gigachadai•17m ago•0 comments

Effective Vibe Coding (Determinism)

https://www.stevenathompson.com/effective-vibe-coding-best-practices-useful-tools/
1•StevenThompson•18m ago•0 comments

Show HN: Post-Interface Design – A manifesto on the abolition of UI

2•andreabergonzi•22m ago•3 comments

Sometimes giving syndication feed readers good errors is a mistake

https://utcc.utoronto.ca/~cks/space/blog/web/FeedReaderErrorsProblemII
1•LorenDB•23m ago•0 comments