研究内容为英文. 本页的研究内容（标题、简报、来源）以英文撰写，未经机器翻译。只有界面已翻译。

新想法Developer tools

"Is it AI?" check API

3周前 · 评分于 2026年10月23日 · ID HBV-261023-09

暂定标题： "Is it AI?" check API and MCP tool for agents before they post, cite or buy (working name)

摘要

简报已验证假设

New / agent-services idea (hunch). Benchmark: Imbue Bouncer open models (imbue-ai/bouncer, AGPL-3.0, 403 stars, Verified GitHub API; browser-confirmed on github.com) and commercial detection APIs (Hive, Sightengine). Existence check: Detection APIs exist (Hive, Sightengine, AI or Not, Reality Defender) and Bouncer rele…

阅读其余部分

ased open models; no product found positioned specifically as a pay-per-call MCP tool for agents with C2PA + detector + reverse-search in one verdict — the gap is packaging only. X: 211 posts/7d (Up 114% Sat→Thu (full UTC days: Sat 17 14 → Thu 22 30); peak Mon 19 80; Thu 22 is 62% below peak — ≥50% below peak: POP 5.3 window-closed signal; window 2026-10-16 10:05–10:20 UTC (start varies by query) → 2026-10-23 10:00 UTC (X API 7-day limit; Fri 16 and Fri 23 are partial UTC days; full days Sat 17 → Thu 22)). Revenue hypothesis: ~$300–$2K/month at month 6 (HYPOTHESIS).

描述假设

One MCP call returns "likely AI / likely human / unsure" with reasons and C2PA credentials for an image, video frame or text, so agents can avoid reposting, citing or buying AI-made content by mistake. Who: Builders of content, research and shopping agents (indie developers, small agencies) who need a cheap, self-serve check.…

阅读其余部分

Pain: Agents repost or cite AI-generated media and buy listings with AI-faked photos; detection APIs are priced and documented for trust-and-safety teams, not for agents. Monetization: Pay-per-call credits (hypothesis ~US$0.002–0.01 per check); free tier via MCP. Build: 1–2 agent-weeks (MCP server, detector ensemble, C2PA read, credits); stack: Vercel functions; open detector models on a GPU host or a wholesale detection API; C2PA library; Stripe credits; MCP server. GTM: MCP directories; GitHub README; GEO.

目标客户

Builders of content, research and shopping agents (indie developers, small agencies) who need a cheap, self-serve check.

痛点

Agents repost or cite AI-generated media and buy listings with AI-faked photos; detection APIs are priced and documented for trust-and-safety teams, not for agents.

为何是这个分数

分数较低的原因： Commodity detection — margins squeezed by wholesale API costs and open models.

护城河假设

Weak: Ensemble + C2PA packaging is copyable. Only distribution in MCP directories helps. Status: Hypothesis.

需求信号

X 上的帖子（7 天）

211

信号趋势

Up 114% Sat→Thu (full UTC days: Sat 17 14 → Thu 22 30); peak Mon 19 80; Thu 22 is 62% below peak — ≥50% below peak: POP 5.3 window-closed signal

网络需求已验证

X: 211 posts in 7 days, peak Mon 19 (80), Thu 22 is 63% below peak (Verified X API, noisy). Viral demand on the human side is large (see HBV-261023-06), but agent-side demand is not quantified.

市场与竞争对手

TAM

Not sized this run — Scope: AI-content detection APIs. Method: No report fetched. (Basis: n/a)

SAM假设

Not sized — Scope: Agent builders paying per check. Method: HBV assumption. (Basis: Hypothesis)

SOM假设

~$4K–$24K ARR (hypothesis) — Scope: Month-6 run-rate. Method: From revenue hypothesis ×12. (Basis: Hypothesis)

竞争对手

Hive AI-generated content detection API; Sightengine; AI or Not API; Reality Defender; Resemble AI (deepfake audio); Imbue Bouncer open models; C2PA verify tools.

商业模式

变现方式假设

Pay-per-call credits (hypothesis ~US$0.002–0.01 per check); free tier via MCP.

收入假设（每月）假设

~$300–$2K/month at month 6 (HYPOTHESIS).

待完成的任务

When my agent is about to post or cite an image, I want to check in one call whether it is likely AI, so I can avoid spreading fakes under my brand.

HXC

Indie dev running a news-summary agent (illustrative) · Agent posts daily image roundups on X; got burned by an AI fake. Buying signal: Would add a cheap MCP check; would not sign an enterprise contract.

构建与 GTM

技术栈

Vercel functions; open detector models on a GPU host or a wholesale detection API; C2PA library; Stripe credits; MCP server

GTM 渠道

MCP directories; GitHub README; GEO

风险

主要风险

Commodity detection — margins squeezed by wholesale API costs and open models.

全部风险

Value: High — Crowded detection APIs | Usability: Attention — Accuracy limits | Feasibility: Attention — Unquantified agent demand | Business Viability: Attention — GPU cost

下一步验证

Do not build now. If HBV-261023-06 (consumer checker) is ever built, expose its backend as an MCP tool at near-zero extra cost.

来源

ideanewagent-servicesai-detectionmcpapideveloper-tools

想在评分当天收到下一个想法吗？

分数、来源、风险和下一步，通过邮件发送。

ProductBots.ai

喜欢这个想法吗？用 ProductBots.ai 把它变成计划。

ProductBots.ai 是一个 Agentic Product OS：由智能体与你一起完成产品工作，从发现到计划。下载这个想法的简报，作为起点。

打开 ProductBots.ai

简报文件为英文。

预览简报

# Briefing: "Is it AI?" check API and MCP tool for agents before they post, cite or buy (working name)

Source: Startup Ideas HQ (HB Ventures), found 2026-10-23. ProductBots Opportunity Score: 55/100, evidence confidence 45%.
Evidence labels (Verified, Self-reported, Hypothesis) are kept as written by the researcher.

## Goal for ProductBots.ai
Turn this idea into a validated MVP plan and a first build. Start from the next validation step; do not assume unlabelled numbers are verified.

## What it is
One MCP call returns "likely AI / likely human / unsure" with reasons and C2PA credentials for an image, video frame or text, so agents can avoid reposting, citing or buying AI-made content by mistake.

New / agent-services idea (hunch). Benchmark: Imbue Bouncer open models (imbue-ai/bouncer, AGPL-3.0, 403 stars, Verified GitHub API; browser-confirmed on github.com) and commercial detection APIs (Hive, Sightengine). Existence check: Detection APIs exist (Hive, Sightengine, AI or Not, Reality Defender) and Bouncer released open models; no product found positioned specifically as a pay-per-call MCP tool for agents with C2PA + detector + reverse-search in one verdict — the gap is packaging only. X: 211 posts/7d (Up 114% Sat→Thu (full UTC days: Sat 17 14 → Thu 22 30); peak Mon 19 80; Thu 22 is 62% below peak — ≥50% below peak: POP 5.3 window-closed signal; window 2026-10-16 10:05–10:20 UTC (start varies by query) → 2026-10-23 10:00 UTC (X API 7-day limit; Fri 16 and Fri 23 are partial UTC days; full days Sat 17 → Thu 22)). Revenue hypothesis: ~$300–$2K/month at month 6 (HYPOTHESIS).

## Customer and pain
Target customer: Builders of content, research and shopping agents (indie developers, small agencies) who need a cheap, self-serve check.
Pain: Agents repost or cite AI-generated media and buy listings with AI-faked photos; detection APIs are priced and documented for trust-and-safety teams, not for agents.

## Market and competitors
TAM: Not sized this run — Scope: AI-content detection APIs. Method: No report fetched. (Basis: n/a)
SAM: Not sized — Scope: Agent builders paying per check. Method: HBV assumption. (Basis: Hypothesis)
SOM: ~$4K–$24K ARR (hypothesis) — Scope: Month-6 run-rate. Method: From revenue hypothesis ×12. (Basis: Hypothesis)
Competitors: Hive AI-generated content detection API; Sightengine; AI or Not API; Reality Defender; Resemble AI (deepfake audio); Imbue Bouncer open models; C2PA verify tools.
Geography: Global (English API docs first); Portuguese docs later

## Business model
Monetization: Pay-per-call credits (hypothesis ~US$0.002–0.01 per check); free tier via MCP.
Revenue hypothesis (monthly): ~$300–$2K/month at month 6 (HYPOTHESIS).

## Build and go-to-market
Build effort (agent-weeks): 1–2 agent-weeks (MCP server, detector ensemble, C2PA read, credits)
Time to first revenue: ~1 month
Tech stack: Vercel functions; open detector models on a GPU host or a wholesale detection API; C2PA library; Stripe credits; MCP server
GTM channels: MCP directories; GitHub README; GEO

## Risks and next step
Main risk: Commodity detection — margins squeezed by wholesale API costs and open models.
Next validation action: Do not build now. If HBV-261023-06 (consumer checker) is ever built, expose its backend as an MCP tool at near-zero extra cost.

## Sources
https://github.com/imbue-ai/bouncer; https://thehive.ai; https://sightengine.com; https://www.aiornot.com; https://www.realitydefender.com; X API counts (window above)

新想法Developer tools· 0周前
AgentParity — continuous parity
A CI service that replays a real agent trajectory against every self-hosted framework release and asserts interrupt/stream/checkpoint/durability parity, failing the build before a regression reaches production — self-serve, $35/repo/month.
64/100
可信度 57%
分数较低的原因： Adjacent platforms (LangSmith, AgentOps, promptfoo) ship parity checks for free
打开 ProductBots.ai →
新想法Developer tools· 1周前
Embeddable 2.5D avatar player SaaS
Hosted drop-in 2.5D lip-sync avatar widget for sites (docs, support, marketing) with API/seat billing — player infrastructure, not a creation studio.
60/100
可信度 53%
分数较低的原因： Owner LivLiv already monetizes hosting and a partner Web SDK; thin whitespace
打开 ProductBots.ai →
现有业务Developer tools· 1周前
NoMac-style copycat
NoMac.app rents disposable Apple-silicon Macs to AI agents by the minute (US$0.90/h Verified list price); TrustMRR shows about US$2,617 revenue in 30 days (+1,217%). HBV angle: a NoMac-style copycat with LATAM billing and distribution.
59/100
可信度 54%
分数较低的原因： Apple licence + hardware economics (fleet cost vs utilization unknown; sub-24h terms Hypothesis) and crowded category
打开 ProductBots.ai →

这是研究，不是投资建议。数字保留其发布时的标签。

摘要

简报✓已验证○假设

描述○假设

目标客户

痛点

为何是这个分数

护城河○假设

需求信号

X 上的帖子（7 天）

信号趋势

网络需求✓已验证

市场与竞争对手

TAM

SAM○假设

SOM○假设