# [012] The opposite side: security scanning for AI coding extensions / Agent Skills / MCP. The event is happening, but supply is free open-source from big companies, so indie developers get no revenue here.

> Source: https://stringers.io/en/o/012.html (Stringers / 什么值得做, generated 2026-11-08)
> Stringers answers one question every day: what small product can one or two people build in a few months and actually make money from. Numbers come from 15 databases we collect daily, each with source and date.
> For agents: these verdicts and the underlying databases (verified revenue, App Store / Google Play charts, ad libraries, GitHub, Product Hunt…) are queryable live over MCP — tools today / hot / products / directions / direction / data. Setup: https://github.com/komako-workshop/stringers (the user logs in at stringers.io → "配置 agent" to get a personal URL). Site index: https://stringers.io/llms.txt

> Every poisoning incident makes "build a scanner" look like an opportunity

[Home](https://stringers.io/worth/) / [012]

# Not promising: security scanners for AI coding extensions, Agent Skills, and MCP. The incidents are real, but big companies give the supply away free and open source, so indie devs make no money here.

Closed Trackable (the opposite side) · looks hot, but it relies on things small teams don't have (enterprise sales, security brand) First spotted 2026-11-01 · last updated 2026-11-02

Status

Closed (not suited for a small team to build alone)

Why this is written up

Every poisoning incident makes "build a scanner" look like an opportunity. But indie devs won't pay for it; enterprises want EDR or MDM integration plus a brand, and big companies already open-source scanners for free.

Confidence

Medium-high. What would overturn this: an indie extension or Skills scanner product hitting over $5k in revenue within 30 days, or the Cursor and VS Code marketplaces starting to require third-party security certification, which would create demand for paid audits.

## Evidence

Grouped by store. Each group starts with the charts that can be graphed (real curves, data dates below the chart), then the scanner's original notes and source links.

### Hacker News2 evidence items

- HN (UTC date): 10-30 "Backdoor in Claude Code for VS Code extension on the Cursor Marketplace" 946 points, 540 comments, best rank #23 on the front page (id 50300683). The phorion.io post says two versions 2.1.312 and 2.1.313 were backdoored with SSH access. It leaves persistence on macOS, Linux and Windows, and sends back environment variable names containing KEY / TOKEN. Cursor pulled it and issued a notice. Same day: "Backdoors in official Claude Code extensions" 655 points, and "Northflank got breached" 461 points. Similar stories in the past 12 weeks: the Rust crate Arrayref poisoning 554 points (August), and Zod 4.6.6 getting poisoned 183 points (October).

- HN: 10-15 Citizen Lab "AdFection: Targeted ads turned into a pipeline for zero-click mercenary spyware", 543 points, 690 comments (id 50101813). Only 10 days passed between the security incident and a detection tool being built. The tool picked up stars but no revenue, which supports this record's point that indie devs can at most ship a free investigation tool as a lead magnet.

### GitHub3 evidence · 8 products with curves

[NVIDIA/SkillSpector](https://github.com/NVIDIA/SkillSpector)

GitHub · created 2026-03-21 · 24,786 stars · 7 days +1,099

Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and supply-chain risks in

Stars

Data 2026-09-01 → 2026-11-08 (69 days, 21 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

[Tencent/AI-Infra-Guard](https://github.com/Tencent/AI-Infra-Guard)

GitHub · created 2024-12-25 · 7,424 stars · 7 days +177

A full-stack AI Red Teaming platform securing AI ecosystems via Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation.

Stars

Data 2026-07-26 → 2026-11-08 (106 days, 13 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

[nolabs-ai/nono](https://github.com/nolabs-ai/nono)

GitHub · created 2026-01-31 · 4,887 stars · 7 days +102

agent runtime security - zero trust, zero setup, zero latency agent sandboxing

Stars

Data 2026-08-02 → 2026-11-08 (99 days, 16 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

[cisco-ai-defense/skill-scanner](https://github.com/cisco-ai-defense/skill-scanner)

GitHub · created 2026-01-29 · 2,701 stars · 7 days +27

Security Scanner for Agent Skills

Stars

Data 2026-10-14 → 2026-11-08 (26 days, 13 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

[perplexityai/numbat](https://github.com/perplexityai/numbat)

GitHub · created 2026-07-24 · 1,162 stars · 7 days +4

Visibility into AI agent activity on endpoints, with on-device detection, optional pre-action blocking, and forensic reconstruction.

Stars

Data 2026-07-30 → 2026-11-08 (102 days, 27 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

[interplains/Zero-Breach-Toolkit](https://github.com/interplains/Zero-Breach-Toolkit)

GitHub · created 2026-10-25 · 4,548 stars · 7 days +555

Free, open-source check for traces of mercenary spyware (Pegasus, Predator, Graphite, Candiru and more) on computers, Android phones and iPhone backups. Double-

Stars

Data 2026-10-28 → 2026-11-08 (12 days, 12 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

[interplains/zero-breach](https://github.com/interplains/zero-breach)

GitHub · created 2026-10-18 · 1,216 stars · 7 days +147

Zero Breach — local, read-only scanner that checks a computer for indicators of the AdFection / Candiru ad-delivered spyware campaign. No network, no admin, no

Stars

Data 2026-10-20 → 2026-11-08 (20 days, 20 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

[hardenrun/aif](https://github.com/hardenrun/aif)

GitHub · created 2026-08-01 · 1,866 stars · 7 days +61 · [see product profile](https://stringers.io/worth/p/agent-integrity-foundation-aif.html)

On device Firewall for Coding and Personal Agents

Stars

Data 2026-10-09 → 2026-11-08 (31 days, 31 observations). Points before 10-28 come from deeply / archive backfill, not every day has one

- GitHub (supply, 2026-10-31 stars): NVIDIA/SkillSpector (security scanner for Agent Skills) 23,687; Tencent/AI-Infra-Guard 7,247; nolabs-ai/nono 4,785; cisco-ai-defense/skill-scanner 2,674, mcp-scanner 1,104; perplexityai/numbat 1,158; plus a dozen more similar projects with 500 to 1,000 stars. All free and open source.

- GitHub (11-02 check): interplains/Zero-Breach-Toolkit (offline check for traces of mercenary spyware like AdFection and Candiru on your machine), created 10-25, 4,100 stars; daily gains dropped from +10-29 at 1,566 to +11-02 at 107, a typical spike. Same author's interplains/zero-breach (created 10-18) has 1,099 stars. All free and open source.

- GitHub(PH 档案 harden.md 和 11-02 核):Harden(VizopsAI,据报道 $2M pre-seed,单一来源)的仓库 hardenrun/aif 只有 README、许可证和 issue 模板,没有源码;1,805 星、392 fork,最新 100 个 fork 里 97 个来自 2026-08 同日注册的新号;10-27 以后星数连续 6 天 0 增长(11-02 核)。官网写「SOC 2 Type II audited」,自家 trust 页写「In progress」。个人版「永久免费」,企业版报价。

### Indie product revenue (TrustMRR)1 evidence · 2 products with curves

[VibeDoctor](https://trustmrr.com/startup/vibedoctor) vibedoctor.io

TrustMRR · 30-day revenue $173 (+102%) · MRR $264 · 17 subscriptions

Monthly revenue (verified by payment platform)

2026-05 → 2026-11 (7 pts). Light bars = current month, not yet complete

Daily revenue (UTC day)

Data 2026-09-29 → 2026-11-06 (39 days, 39 observations). Single-day swings are big, so look at the trend in the monthly breakdown above.

[isMalicious](https://trustmrr.com/startup/ismalicious) ismalicious.com

TrustMRR · 30-day revenue $1,556 (-18%) · MRR $1,731 · 14 subscriptions

Monthly revenue (verified by payment platform)

2025-11 → 2026-11 (13 pts). Light bars = current month, not yet complete

Daily revenue (UTC day)

Data 2026-09-29 → 2026-11-06 (39 days, 39 observations). Single-day swings are big, so look at the trend in the monthly breakdown above.

- Indie (30 revenue, as of 10-31): ShipSafe (CLI security scanning for the agent era, $9/mo) $9; VibeDoctor $90; isMalicious (threat intel) $1,558 (−18.6%). No product that scans extensions, Skills, or MCP servers has meaningful revenue.

### Other sources (official sites, press releases, web pages) 1 pieces of evidence

- Supporting evidence: the DSH plugin with the most stars in [010] was already gray (free model access, jailbreak prompts). Trust problems in plugin ecosystems are structural.

## Already being built (11-02 backfill)

Zero-Breach-Toolkit and zero-breach (free detection tools, incident-response type); Harden (funded company, free for individuals, enterprise quote-based, repo has no source code).

## Who it's for

Only suits teams already selling enterprise security. An indie dev could at most ship a free local investigation tool as a lead magnet (for example, a one-click check of ~/.ssh/authorized_keys and LaunchAgents for changes made by extensions).

## Should every product builder know this?

Yes. Anyone building IDE extensions, Skills, or MCP servers should know that distribution channels are under attack and their own extensions could become the attack vector.

## Related product profiles

[Harden(AIF / Agentic Integrity Foundation)](https://stringers.io/worth/p/agent-integrity-foundation-aif.html)

## Change log

2026-11-02

Backfilled (from the "GitHub new opportunities (batch 1)" §1.10 and the "Product Hunt roundup v1" Harden profile): both supporting signals point to the same conclusion. First, "security incident → free detection tool within 10 days → stars, no money" (Zero-Breach-Toolkit: 7 days, 4,000 stars, already fading). Second, the funded company in this space (Harden) is also just giving things away for free to build a funnel, and its GitHub numbers aren't trustworthy yet. Status unchanged.

2026-11-02

ShipSafe $9, VibeDoctor $111 (+29.7%), SkillSpector 23,836 stars. Unchanged.

2026-11-01

First recorded.

Original: Scanner opportunity library #012 (updated daily, 06:00); this page was generated on the 2026-11-08 edition.