HK
Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Top
New
Best
Ask
Show
Jobs
Request
31.
▲
Show HN: First-token-only flaw in Claude Code permissions (triage bot too)
spitfirecowboy.com
discuss
3 months ago
Apylon777
3 points
32.
▲
I built an AI browser with prompt-injection defense at 16 on an i5 with 8GB RAM
discuss
4 months ago
latestinssan
3 points
33.
▲
Show HN: ClawGuardian – Security Plugin for OpenClaw
github.com/superglue-ai
discuss
5 months ago
sfaist
3 points
34.
▲
Show HN: Destructive_command_guard (Dcg)
github.com/Dicklesworthstone
discuss
5 months ago
eigenvalue
3 points
35.
▲
Show HN: I built a sandboxed VM for letting AI agents go wild without risks
discuss
5 months ago
pancakeInDev
3 points
36.
▲
Show HN: Mother May I? – Auto-approve safe Bash commands in Claude Code
github.com/dgerlanc
discuss
5 months ago
dgerlanc
3 points
37.
▲
Show HN: SafeShell – reversible shell commands for local AI agents
github.com/qhkm
discuss
6 months ago
qhkm
3 points
38.
▲
Tell HN: Recent SUSE update caused my first real dataloss in 20 years
5 comments
5 years ago
2ton_jeff
2 points
39.
▲
Show HN: Vigil – Zero-dependency safety guardrails for AI agent tool calls
2 comments
4 months ago
HexitLabs
2 points
40.
▲
Show HN: AI-runtime-guard – Policy enforcement layer for MCP AI agents
github.com/jimmyracheta
2 comments
4 months ago
JimmyRacheta
2 points
41.
▲
Show HN: EdgeAI-OS – Air-gapped Linux distro where AI is a system primitive
2 comments
4 months ago
neuralweaves
2 points
42.
▲
Show HN: Supe – Give your AI agent a brain, not just memory
github.com/xayhemLLC
2 comments
5 months ago
xxayh
2 points
43.
▲
Show HN: LawClaw – Constitutional governance for AI agents (MIT)
1 comment
4 months ago
nghiahsgs
2 points
44.
▲
Show HN: Raypher–eBPF-based runtime security and hardware identity for AI agents
github.com/kidigapeet
1 comment
4 months ago
Kidiga
2 points
45.
▲
Show HN: Rampart – Open-source security for Claude and AI agents in YOLO mode
github.com/peg
1 comment
4 months ago
cl4p
2 points
46.
▲
Show HN: OpenClaw is just not dangerous enough. I needed something else
github.com/kkovacs
discuss
a month ago
kkovacs
2 points
47.
▲
Show HN: Ryvos – Autonomous AI assistant in Rust(15MB RAM,50 tools,16 providers)
ryvos.dev
discuss
4 months ago
aayush-mishraaa
2 points
48.
▲
Show HN: Sidekick – See what your AI coding agents are doing
cesarandreslopez.github.io
discuss
4 months ago
cal_lopez
2 points
49.
▲
Show HN: Rampart – Runtime firewall for Claude Code and AI agents in YOLO mode
github.com/peg
discuss
4 months ago
trevxr
2 points
50.
▲
Show HN: Created a script to run commands in an isolated VM for AI tool calling
github.com/bigattichouse
discuss
a year ago
bigattichouse
2 points
51.
▲
Show HN: A little bash script to install the latest Git (2.41.0)
discuss
3 years ago
graderjs
2 points
52.
▲
Show HN: Assign tasks to 7 AI agents with -mentions, autonomous mode, OpenClaw
2 comments
4 months ago
bahaAbunojaim
1 points
53.
▲
Show HN: Core Rth. A governed AI kernel for engineers who don't trust their LLMs
github.com/rthgit
1 comment
4 months ago
christianrth
1 points
54.
▲
Show HN: Kintsugi – a local-first safety net for AI agents and humans
github.com/arrowassassin
discuss
6 days ago
arr0wassass1n
1 points
55.
▲
Show HN: Nonstop, two files that keep Claude Code working while you sleep
github.com/andylizf
discuss
3 months ago
andylizf
1 points
56.
▲
Execwall – firewall to stop ModelScope CVE-2026-2256 (AI agent command injectn)
discuss
3 months ago
sentra
1 points
57.
▲
Show HN: Gohpts-IPv4/IPv6/TCP/UDP Transparent Proxy with ARP/NDP/Rdnss Spoofing
github.com/shadowy-pycoder
discuss
3 months ago
shadowy-pycoder
1 points
58.
▲
Show HN: Kvlar – Open-source firewall for AI agent tool calls
github.com/kvlar-io
discuss
4 months ago
kvlar
1 points
59.
▲
Show HN: Nexus AI – DevOps terminal bridge to Telegram (local-first, BYOK)
remote-nexus.dev
discuss
4 months ago
BSanroma
1 points
60.
▲
Show HN: DevSweep – A TDD-backed CLI to clean artifacts safely
github.com/Sstark97
discuss
5 months ago
ascinfo
1 points
More