prompt on Hacker News

82 titles from 80 authors in the last 30 days, with 80 distinct links and 20 Show HN.

Counts titles, it does not confirm facts. Open the source before concluding anything.

Projects on this term

The score comes from our model.

• Show HN: My personal agent, distributed as a prompt · score 7.0

  life is a self-hosted AI personal assistant built on Claude Code that runs agents on a Mac, accessible via iPhone, desktop, and web over Tailscale. It enables editable apps, goal tracking, approvals, and data integration, all while keeping data private. The project's innovative distribution via a single prompt and self-modifying app concept make it interesting.

• Show HN: Kompressor – deterministic prompt compression for coding agents · score 7.0

  OptScale Kompressor is a Rust-based proxy that deterministically compresses tool outputs in coding agent prompts, reducing token costs while preserving prompt-cache compatibility. It achieves 22.7% cost savings on Terminal-Bench 2.1, outperforming similar tools like Headroom, and ensures every removed detail remains retrievable.

• Show HN: Pi subagents Lite – keep your system prompt clean · score 6.8

  pi-subagents-lite is a TypeScript plugin for the pi AI coding assistant that enables spawning isolated sub-agents with separate tools, extensions, and models. It focuses on minimal token overhead via compact tool schemas and offers features like background agents, steering, worktree support, and a live UI widget. It's interesting for its schema-first design and practical multi-agent orchestration within a CLI context.

• Show HN: Chrome extension that types one prompt into many AI chat tabs · score 6.7

  A Chrome extension that broadcasts a single prompt to multiple AI chat tabs (e.g., ChatGPT, Claude, Gemini) simultaneously, enabling side-by-side comparison of responses. It simplifies multi-AI workflows by eliminating manual copying and pasting between tabs.

Recent titles

1. Brazilian lawyers fined $16K for hiding a prompt injection in a court filing
   2026-11-01 · tomshardware.com · HN discussion
2. I recreated subvocal in 9 minutes with one prompt for AirPods Pro 3
   2026-11-01 · twitter.com · HN discussion
3. Brazilian Court Sanctions Lawyers for Prompt Injection in Court Filing
   2026-11-01 · jetstream.security · HN discussion
4. Prompt caching: same agent, same rules, about a quarter of the bill
   2026-10-31 · blog.getwerk.ai · HN discussion
5. State of the Prompt: What People Ask AI
   2026-10-31 · stateoftheprompt.net · HN discussion
6. See how AI search reads a page and fix it in one prompt
   2026-10-31 · lumi-geo.pages.dev · HN discussion
7. PrimeForge AI – Multi-channel marketing campaign generator from a single prompt
   2026-10-30 · HN discussion
8. Show HN: Prompt – A New, Beautiful Terminal
   2026-10-30 · github.com · HN discussion
9. Show HN: Type and+ in Claude Code and a second Claude rewrites your prompt first
   2026-10-29 · github.com · HN discussion
10. Prompt Layering
    2026-10-29 · agentpatterns.ai · HN discussion
11. Show, don't prompt: how QA Wolf turns a recording into a real test
    2026-10-29 · qawolf.com · HN discussion
12. Tell HN: Claude Code adds your email to the prompt, the LLM sends it to websites
    2026-10-29 · HN discussion
13. ARCR: Prompt Architect
    2026-10-28 · agentowl.netlify.app · HN discussion
14. A safety checklist in the prompt didn't measurably help frontier models
    2026-10-28 · theagentforge.io · HN discussion
15. A prompt marketplace where every prompt is run on its model before sale
    2026-10-27 · gradedprompts.com · HN discussion
16. Blazle – Block prompt injection and secret leaks before AI agents act
    2026-10-27 · blazle.io · HN discussion
17. Prompt Patterns That Compound
    2026-10-27 · patrickmccanna.net · HN discussion
18. Who Governs the Governor? – Stored Prompt Injection Through the API Gateway
    2026-10-27 · druhindhavala.com · HN discussion
19. Tokensaver, see which parts of your LLM prompt cost money
    2026-10-26 · github.com · HN discussion
20. Prompt injection scanners should be positional, not phrasal
    2026-10-24 · attest.rta.lat · HN discussion
21. Pair Prompt and Agentic Interview
    2026-10-24 · liveshortly.com · HN discussion
22. Top AI Prompt Engineering Best Practices for Developers
    2026-10-23 · aiqualityhq.com · HN discussion
23. Show HN: Kuron – A 5-device airline system from one prompt, no code generated
    2026-10-23 · claude.ai · HN discussion
24. Show HN: AI Prompt Chain Builder – offline multi-step prompt workflows
    2026-10-22 · theaioperatorstore.gumroad.com · HN discussion
25. Show HN: AIQualityHQ – Browser-native prompt QA and security guardrails
    2026-10-22 · aiqualityhq.com · HN discussion
26. Prompt "find bad vulnerabilities in PHP" was enough to find the PHP vuln
    2026-10-22 · twitter.com · HN discussion
27. Show HN: My personal agent, distributed as a prompt
    2026-10-21 · github.com · HN discussion
28. Pencil versus Prompt
    2026-10-20 · avanelsas.github.io · HN discussion
29. Building a System One Model for Prompt Injection
    2026-10-20 · stackone.com · HN discussion
30. Show HN: SentinelGate – Drop prompt injections at Token 0 before LLM inference
    2026-10-20 · sentinelgate.dev · HN discussion

All trends