LAUNCH·INDEX
LNCH1,007+7.6%

Experimental: traction = early-adopter votes (PH) / points (HN), not downloads or usage; updated about once a day and frozen ~31–36h after launch on Product Hunt (7 days on Show HN); sectors come from keyword rules (~70–75% accurate); data starts Oct 3 (HN) / Oct 9 (PH). “—” means not enough data; nothing is estimated. How it works

AI Agents (AGNT)

Sector · Show HN · data as of Oct 17, 2026, 12:21 PM UTC

80−9 (−10.1%)

launches · 3D Oct 14–Oct 16 vs Oct 7–Oct 9 (same weekday) (89)

7.7%—

hit rate · Oct 12–Oct 14 settled (7/91) vs Oct 5–Oct 7 (8/104)

— Not significant at current sample size: +0.0pp, 104 → 91 settled launches (needs 100+ in each window).

−5.0

trending score

Launches & hit rate

Supply · launches per day
02550Oct 3: 11 launches (incomplete day)Oct 4: 16 launchesOct 5: 33 launchesOct 6: 35 launchesOct 7: 36 launchesOct 8: 27 launchesOct 9: 26 launchesOct 10: 14 launchesOct 11: 18 launchesOct 12: 16 launchesOct 13: 46 launchesOct 14: 29 launchesOct 15: 27 launchesOct 16: 24 launchesOct 17: 5 launches (incomplete day)Oct 3Oct 5Oct 7Oct 9Oct 11Oct 13Oct 15Oct 17
Demand · hit rate per day (share of settled launches with 10+ votes)
0%10%20%Oct 3: incomplete dayOct 4: 0 of 16 settled launches hit 10+ votes (0.0%)Oct 5: 1 of 33 settled launches hit 10+ votes (3.0%)Oct 6: 3 of 35 settled launches hit 10+ votes (8.6%)Oct 7: 4 of 36 settled launches hit 10+ votes (11.1%)Oct 8: 3 of 27 settled launches hit 10+ votes (11.1%)Oct 9: 2 of 26 settled launches hit 10+ votes (7.7%)Oct 10: 0 of 14 settled launches hit 10+ votes (0.0%)Oct 11: 2 of 18 settled launches hit 10+ votes (11.1%)Oct 12: 2 of 16 settled launches hit 10+ votes (12.5%)Oct 13: 3 of 46 settled launches hit 10+ votes (6.5%)Oct 14: 2 of 29 settled launches hit 10+ votes (6.9%)Oct 15: launches not settled yetOct 16: launches not settled yetOct 17: incomplete dayOct 3Oct 5Oct 7Oct 9Oct 11Oct 13Oct 15Oct 17

4 days left blank: votes still coming in (launches under 48h old), incomplete day, or no launches — not shown as zero.

Faded bars are incomplete days (today, or days before every selected source was being collected).

Key statistics

Launches
24Oct 16
Avg daily launches
26.713d
7D range
14 – 46Oct 10–Oct 16
Hit rate
8.0%14/176 · Oct 8–Oct 14
Volatility
67%12 daily chg
Share of market
17.8%7D
Top source
Show HN100%
Previous window
89Oct 7–Oct 9

Stocks in this sector (AGNT)

Highest traction among products launched in the last 7 days. Traction for older Product Hunt launches is frozen (votes stop being re-read ~31–36h after launch; Show HN after 7 days), so this is each product’s last recorded traction. 172 matches. Spam giveaways excluded.

#Sector
1SolidCraft

Open-Source Parametric CAD in Rust (With MCP Support)

AGNT14100.0%04d ago
2An experimental pattern-first DAW that supports MCPAGNT8400.0%04d ago
3rStation

A shared research lab for your AI agents

AGNT53+3▲ +6.0%030h ago
4Graphicle, a Rust port of Graphify with memory for coding agentsAGNT4700.0%03d ago
5MemorySync

Memory for AI Agents

AGNT2200.0%04d ago
6Jevcrawl

A web crawler for AI agents with no LLM in the crawl loop

AGNT2100.0%03d ago
7Plimsoll, a self-hosted sandbox for code your AI agent writesAGNT1600.0%05d ago
8BlastLoop, a structure-first coding method for AI agentsAGNT1500.0%02d ago
9Littleguys

Claude Code, Codex and OpenClaw agents in your menu bar

AGNT1500.0%006d ago
10Just gave Claude Code the website and ad budget to see what it would doAGNT1100.0%02d ago
11AxiomCode Graph, Reducing coding agent rework with formal methodsAGNT1000.0%03d ago
12Skills Hub, a desktop app for managing AI agent skillsAGNT1000.0%05d ago
13E2E

Open-source agentic testing framework for web, iOS and Android

AGNT900.0%05d ago
14Weatherdeck

a read-only status board for AI coding agent lanes

AGNT9+1▲ +12.5%0026h ago
15Chiplab

Give your AI agent access to virtual hardware

AGNT800.0%03d ago
16IHMT

Infinite Hierarchical Memory Tree for AI Agents

AGNT800.0%05d ago
17RSI-Anything

JEPA world models for self-improving agents

AGNT800.0%004d ago
18GSD Task Manager

an MCP server that gives your AI agent a task list

AGNT700.0%06d ago
19I Cut my Agent Tokens By 50%AGNT600.0%06d ago
20OpenWants

A Simulated City Where AI Agents Handle Residents Needs

AGNT6frozen——7d ago
21Let Claude Code answer with a diagram page instead of a wall of textAGNT600.0%04d ago
22RDST

A local agent that watches Postgres/MySQL for query regressions

AGNT600.0%02d ago
23RL agent trained only in synthetic markets got better on real 2020crashAGNT600.0%004d ago
24Play a 5 minutes match of CS 1.6 1v1 while you wait for your agentAGNT500.0%023h ago
25Shrine, a calm pixel room on your Mac for all your coding agentsAGNT500.0%03d ago
26I made an agent can self-improving while workingAGNT500.0%05d ago
27Pi subagents Lite

keep your system prompt clean

AGNT500.0%02d ago
28Recover Claude Code Desktop sessions after switching accountsAGNT500.0%04d ago
29Evaluate and improve your agent steeringAGNT40—03d ago
30Rezoom

one chat list for all your Claude Code sessions

AGNT40—06d ago
31server-noonien

conflict-free agent memory across machines, no server

AGNT40—02d ago
32Ramet

Persistent, forkable workspaces for AI agents

AGNT40—05d ago
33CoolDock

a second Dock for macOS with live widgets and an MCP server

AGNT40—002d ago
34Autograph

a music notation editor you can drive with Claude over MCP

AGNT40—004d ago
35Benchmark coding agents on deploying appsAGNT30—03d ago
36I built Airlock, to sandbox your AI agent and know what it's doingAGNT30—05d ago
37PTrace

An iPhone Running Tracker with MCP

AGNT30—033h ago
38Herdr for Android

The runtime coding agents run on

AGNT30—13d ago
39Open-source Mac menu bar app for Claude, Codex and CursorAGNT30—04d ago
40shadowclone, skills learned from your coding agent correctionsAGNT30—04d ago
41Deterministic AI Agents GuardrailAGNT30—0033h ago
42Sugabots

Group chats where your team and AI agents work together

AGNT30—003d ago
43Fresh cleanup and reset a project after coding agent prsAGNT30—003d ago
44Personal Agent Proxy (Open Source)AGNT30—004d ago
45Laya guard

a local check before your coding agent runs a command

AGNT30—004d ago
46Stepbook

run long agent tasks as short steps with saved state

AGNT30—004d ago
47A calendar tracker for OpenAI's 28-day Codex reset challengeAGNT30—006d ago
48An agent harness for long-running pentests and code auditsAGNT30—007d ago
49An OpenAI-compatible API that coding agents sign up for themselvesAGNT20—05d ago
50Ret

An inbox and canvas for your Codex and Claude Code agents

AGNT20—04d ago

Traction = current votes (PH) / points (HN) — early-adopter attention, not downloads or usage. Gained = traction gained between the two latest daily ingest runs; Gain % only when the previous traction was 5+. Buzz = comments gained in that interval. “frozen” = no longer tracked, so no current gain. How it works

AGNT · AI Agents · Launch Index