The security layer for AI-generated code
Make AI-generated code safer to ship.
Verglos is a local-first security layer between code change and release: scan locally, decide with explicit policy, and prepare bounded evidence for handoff. The free local scanner is live today; the v2 track adds local Hunt verification and signed evidence as each path qualifies.
$ npx verglos
scanned 1,247 files in 42s
report: verglos-report.html + verglos-report.json
critical / D4-004
credential-shaped value in source
high / AI-005
unresolved package name in lockfile
fix
move to env, rotate, pin dependency
Critical risk
2.0.0-alpha.1
latest npm
public qualification prerelease; stable scanner 1.8.3
10
coverage domains
explicit JS/TS detector map
0
source upload
scan runs on your machine
A1/A2
Hunt assurance
qualification-gated bounded paths
Why the market changed
Agentic AppSec is now table stakes.
ZeroPath, Corgea, Snyk Agent Fix, Cycode Maestro, XBOW, Semgrep AI Assistant, and Pixee all point at the same market shift: scanners are becoming triage-and-fix agents.
The shared pattern is cloud LLM triage plus PR fixes. That is useful, but it still asks the buyer to trust inferred severity and generated remediation.
Verglos takes a different lane: local sandbox proof, AI-authorship provenance, and signed evidence artifacts. It is built to prove what fired, show where agent-authored code concentrated risk, and package the result for handoff.
The Verglos flow
One release decision. Every proof point connected.
Verglos keeps source and scanning local, then connects its findings with evidence from Trivy, Docker/OCI, and the tools your team already runs. The result is a decision you can explain, not another isolated alert.
Hunt and signed records are shown with their declared assurance and qualification boundary.
Inspect
Scan your JS/TS code
Findings, provenance, score, and local reports.
Decide
Correlate + policy
Normalize evidence into pass, review, or block states.
Prove
Bounded verification
Approved local Hunt and portable records where qualified.
Monitor
Remember the decision
Hosted history, alerts, and team context remain qualification-gated.
Customer-controlled execution: code, approved fixtures, scans, and ordinary reports stay local.
Qualification boundary: Hunt proof and portable records disclose their assurance and limitations.
Optional hosted memory: history, monitoring, alerts, and team context remain plan- and qualification-gated.
SAST + SCA · 10 domains
Built from the bugs AI agents make cheap to create.
One CLI runs SAST (code-side detectors — secrets, injection, misconfig, AI patterns) and SCA (dependency CVEs, vendored libraries, slopsquat / typosquat) across ten explicit domains: injection, auth, access control, cryptography, misconfiguration, supply chain, API surface, data exposure, runtime, and monitoring.
Domain 01 of 10
Injection & input flaws
Critical
Every piece of user input is a weapon if it reaches a dangerous sink unsanitised. We trace source-to-sink across your codebase — SQL, NoSQL, command exec, XSS, SSTI, path traversal — and flag the concatenations, spreads, and eval calls that turn a request into a takeover.
sample finding
element.innerHTML = req.query.name
app/user/[id]/page.tsx:42
Domain 01 of 10
Injection & input flaws
Every piece of user input is a weapon if it reaches a dangerous sink unsanitised. We trace source-to-sink across your codebase — SQL, NoSQL, command exec, XSS, SSTI, path traversal — and flag the concatenations, spreads, and eval calls that turn a request into a takeover.
What Verglos catches
- SQL / NoSQL
- Command exec
- XSS (stored / reflected / DOM)
- SSTI
- Path traversal
- XXE
Sample finding
⚠ CriticalDomain 01
element.innerHTML = req.query.name
→ app/user/[id]/page.tsx:42
Three moats
Verglos is not trying to win the cloud triage race.
AI-authorship provenance
Every finding is annotated human / AI / mixed with commit-shape and agent-artifact evidence.
Cloud scanners can triage code, but they usually do not preserve authorship evidence as a first-class risk dimension.
Local-first
Scan runs 100% on the developer's machine today. Hunt is designed for local verification so source does not need to leave the machine.
Cloud LLM products inherit cloud cost and source-upload friction; Verglos keeps the security loop local.
Signed evidence artifact
The v2 attest track is designed around a portable JSON+HTML bundle signed with Ed25519; hosted signing and public verification remain qualification-gated.
A PR fix is not a client-portable evidence artifact. Studio is built around the handoff surface.
Evidence path
Enterprise thinking without pretending to be enterprise theater.
Verglos is designed to move from local findings to durable proof: CI enforcement, monitoring, signed attestations, verification URLs, SBOMs, and client-ready reports.
01
Inspect
Run the local JS/TS scanner and keep findings, provenance, and ordinary reports beside your code.
02
Decide
policy-bound
Normalize evidence and apply explicit policy so pass, review, and block states stay explainable.
03
Prove
qualification-gated
Use approved bounded Hunt paths and portable records only where the declared runtime and evidence contract support them.
04
Monitor
hosted qualification
Carry qualified decisions into optional history, alerts, and team context without implying hosted coverage is live everywhere.
WHAT VERGLOS IS NOT
Not another autonomous pentester. Verglos verifies findings on code you own.
Not a cloud SAST. Verglos does not upload source to run an LLM triage pass.
Not a Vanta replacement. Verglos produces evidence artifacts, not the compliance program.
Not AI-powered as a headline claim. The scanner is deterministic; the hunt loop uses local synthesis and sandbox execution.
Why it matters
The invisible cost you're paying today.
Not the bugs Verglos finds — the ones you don't. Every AI-assisted stack carries a handful of buried failure modes that won't announce themselves. Here's what changes when you can see them.
₹2,809.43/mo · less than one hour of dev time
The developer
Know what changed before it ships.
Run the local scanner against the project you control, inspect findings and provenance, and keep the HTML/JSON result with the change.
The agency
Show the handoff boundary.
When Studio custody is qualified, send an approved redacted projection with its digest and limitations instead of exposing source or raw findings.
The founder
Keep release decisions explainable.
Policy and record surfaces are designed to preserve why a result was accepted, reviewed, or blocked; hosted history and monitoring remain qualification-gated.
The agent user
Put authority around agent actions.
The CLI and MCP contracts bind supported actions to explicit inputs, approvals, and bounded outputs; arbitrary model-generated shell is not accepted.
Language coverage
JavaScript and TypeScript first.
JS/TS is live today because that is where Verglos has real detector coverage, package analysis, and provenance signal. Other stacks stay on the request list until they are actually supported.
Live now
JavaScript and TypeScript first.
Covers Node.js, Next.js, React, Express, Fastify, and npm workspaces.
Available today
LiveCovers Node.js, Next.js, React, Express, Fastify, and any npm / pnpm / yarn workspace. Ten detection domains — secrets, injection, dependency CVEs, misconfigurations, git history, and more.
Requested next
ComingWant your stack prioritised? Tell us which one you ship on →
npm
Install in one command.
The package is the fastest path to value: no signup, no install ceremony, no source upload.
GitHub
Inspect the scanner.
Technical reviewers can inspect the CLI, detectors, packages, docs, release notes, and evidence docs.
Top Notchh
Backed by a product company.
Verglos is part of a focused product portfolio, not a one-off landing page or loose security script.
Plans
Free is the complete local scanner.
Run it as many times as you want locally. Paid tiers are for teams that need enforcement, monitoring, evidence, handoff, and readiness workflows around the scan.
Free plan
Unlimited local scans. Full findings. Full paths. Local HTML and JSON reports. No account required.
Live
Free
$0
For: Developers, founders, agencies
Use: OSS scanner, provenance, and MCP scan.
Unlimited local scans. No account, no source upload, no paywall for findings, paths, or line numbers.
Catalog terms
- Billing
- $0 · Live
- Seats
- 1 seat
- Applications
- Not set in public catalog
- Records
- Not set in public catalog
- Client workspaces
- Not set in public catalog
- Hosted retention
- 0 hosted days
- Public verification requests
- Not set in public catalog
- Managed execution runs
- Not set in public catalog
- Overages
- No overage schedule published
- /Full SAST + SCA scan across 10 domains
- /Security score /100
- /All file paths + line numbers
- /SAST: secrets, entropy, injection, misconfig, AI patterns
- /SCA: dependency CVEs via OSV + vendored-library CVE cross-ref
- /AI-authored code detection
- /Slopsquat / hallucinated-package check
- /Local HTML + JSON report
- /Pre-commit hook
- /MCP verglos_scan tool
- /CI blocks on criticals
- /Unlimited local use
Live
Pro
₹2,809.43/mo
For: Serious solo devs and small teams
Use: Hunt track, CI discipline, and agent guardrails.
For developers whose coding agent should verify high-signal findings before every commit with approved, bounded recipes. Hosted entitlement and final acceptance remain qualification-gated.
Catalog terms
- Billing
- ₹2,809.43/mo or ₹28,094.3/yr · Live
- Seats
- 2 seats
- Applications
- 5 applications
- Records
- 100 records/month
- Client workspaces
- Not set in public catalog
- Hosted retention
- 90 hosted days
- Public verification requests
- Not set in public catalog
- Managed execution runs
- Not set in public catalog
- Overages
- No overage schedule published
- /Everything in Free
- /verglos hunt bounded A1/A2 path for Critical + High
- /MCP verglos_hunt_finding (approval-limited)
- /MCP verglos_hunt_report (approval-limited)
- /MCP verglos_hunt_before_write (approval-limited)
- /MCP verglos_hunt_explain_verdict (approval-limited)
- /verglos fix — provenance-annotated PR diffs
- /verglos ci --hunt path for verified criticals
- /CI score thresholds
- /Continuous CVE monitoring
- /Offline-safe license (Ed25519 signed)
Validation
Team
Not purchasable
For: Growing engineering teams
Use: Shared policy and CI decision workflows.
The Team catalog tier is defined for the shared-policy surface and remains unavailable for new purchase while that hosted path is being qualified.
Catalog terms
- Billing
- Not purchasable · Validation
- Seats
- 5 seats
- Applications
- 25 applications
- Records
- 500 records/month
- Client workspaces
- Not set in public catalog
- Hosted retention
- 365 hosted days
- Public verification requests
- Not set in public catalog
- Managed execution runs
- Not set in public catalog
- Overages
- No overage schedule published
- /Everything in Pro
- /Shared organization policy
- /Repository-bound CI decisions
- /Team seats and application allowances
Validation
Studio
Not purchasable
For: Agency owners and delivery leads
Use: Attestation, client handoff, and proof resale.
Agencies do not need another noisy scanner. They need signed evidence they can hand to clients.
Catalog terms
- Billing
- Not purchasable · Validation
- Seats
- 10 seats
- Applications
- 100 applications
- Records
- 2,000 records/month
- Client workspaces
- 20 client workspaces
- Hosted retention
- 1,095 hosted days
- Public verification requests
- Not set in public catalog
- Managed execution runs
- Not set in public catalog
- Overages
- No overage schedule published
- /Everything in Pro
- /verglos hunt on Medium
- /Portable evidence records — qualification-gated
- /Public verify URL — qualification-gated
- /White-label report — qualification-gated
- /Firecracker sandbox adapter — qualification-gated
- /Agency dashboard — qualification-gated
- /MCP verglos_attest
Contact sales
Enterprise
Custom
For: Security-led organizations
Use: Enterprise controls around the evidence chain.
For teams that need organization-level administration, private verification infrastructure, and custom detector coverage.
Catalog terms
- Billing
- Custom · Contact sales
- Seats
- Contracted
- Applications
- Contracted
- Records
- Contracted
- Client workspaces
- Contracted
- Hosted retention
- Contracted
- Public verification requests
- Contracted
- Managed execution runs
- Contracted
- Overages
- No overage schedule published
- /Everything in Studio
- /SSO / SCIM
- /Self-hosted verify chain
- /Audit log
- /Custom detector packs
| Capability | Free | Pro | Team | Studio | Enterprise |
|---|---|---|---|---|---|
| verglos scan (all detectors) | Yes | Yes | Yes | Yes | Yes |
| Provenance layer + slopsquat + typosquat | Yes | Yes | Yes | Yes | Yes |
| Local HTML + JSON report | Yes | Yes | Yes | Yes | Yes |
| MCP verglos_scan tool | Yes | Yes | Yes | Yes | Yes |
| verglos ci blocks on any critical | Yes | Yes | Yes | Yes | Yes |
| Pre-commit hook | Yes | Yes | Yes | Yes | Yes |
| verglos secrets / deps / score | Yes | Yes | Yes | Yes | Yes |
| verglos hunt on Critical + High | No | Qualification-gated A1/A2 | Qualification-gated A1/A2 | Qualification-gated A1/A2 | Qualification-gated A1/A2 |
| verglos hunt on Medium | No | No | No | Qualification-gated | Qualification-gated |
| MCP hunt tools | No | Qualification-gated A1/A2 | Qualification-gated A1/A2 | Qualification-gated A1/A2 | Qualification-gated A1/A2 |
| MCP verglos_hunt_before_write | No | Qualification-gated | Qualification-gated | Qualification-gated | Qualification-gated |
| verglos fix (provenance-annotated PR diffs) | No | Yes | Yes | Yes | Yes |
| verglos ci --hunt | No | Qualification-gated | Qualification-gated | Qualification-gated | Qualification-gated |
| CI score threshold | No | Yes | Yes | Yes | Yes |
| Continuous CVE monitoring | No | Yes | Yes | Yes | Yes |
| verglos attest (Ed25519 signed bundle) | No | No | No | Qualification-gated | Qualification-gated |
| Public verify URL | No | No | No | Qualification-gated | Qualification-gated |
| White-label report | No | No | No | Qualification-gated | Qualification-gated |
| Firecracker sandbox adapter | No | No | No | Qualification-gated | Qualification-gated |
| Agency dashboard | No | No | No | Qualification-gated | Qualification-gated |
| SSO / SCIM · self-hosted verify · audit log · custom detectors | No | No | No | No | Contact sales |
Top Notchh operating stamp
Built by a product company, not a scanner content farm.
Top Notchh Solutions builds practical software products from real operational pain. Verglos is the security decision-layer product in that portfolio: specific, local-first, honest about maturity, and built to become infrastructure for AI-assisted delivery.
From the Journal
Research and product notes.
Original research from the Verglos scan campaign, positioning notes, and product decisions — published on the Top Notchh Journal.
Read the full Journal →Research · scan campaign
2026-07-30
We scanned 300 world-class TypeScript & JavaScript repos. 57.7% had critical or high findings.
Verglos ran the ICP-relevant top 300 GitHub repositories. 169 of 293 successful scans surfaced critical or high risk — the finding that shaped the product.
Read on Top Notchh →Patterns · developer
2026-07-30
The security bugs AI code keeps writing: 8 patterns from the ICP top 300.
SQL injection, weak tokens, IDOR, wildcard CORS, verbose stack leaks, hard-coded secrets, mass assignment, and slopsquat risk — the AI-era patterns that repeat across the campaign.
Read on Top Notchh →Positioning · comparison
2026-07-25
Verglos vs Snyk, GitHub, Semgrep, Socket, and Gitleaks: why AI-built software needs a different layer.
A founder-friendly comparison against the top scanners and why AI-built software needs evidence, not more alerts.
Read on Top Notchh →Thesis · agencies & founders
2026-07-24
The security decision layer for AI-built software.
AI-assisted teams do not just need scanners. They need a defensible decision layer that turns findings into bounded evidence for clients, buyers, and auditors. The business case for Verglos.
Read on Top Notchh →Disclosure · secrets
2026-07-19
How Verglos handles active credential disclosures.
A responsible-disclosure note on live credentials, private maintainer contact, and why public proof should wait until rotation is confirmed.
Read on Top Notchh →
FAQ
Boundaries make the product stronger.
Is Verglos just another scanner?+
No. The scanner is the entry point. Verglos is built around evidence: AI-code risk signal, report artifacts, MCP guardrails, CI enforcement, and client-ready proof.
Does Verglos upload source code?+
The local scanner writes HTML and JSON reports beside your code. Networked flows are explicit: login, activation, monitoring registration, and future hosted evidence.
Which languages are supported today?+
JavaScript and TypeScript are live. Other language icons are request-list indicators, not shipped coverage.
Who is behind Verglos?+
Verglos is built by Top Notchh Solutions, an AI-native product company building practical software products from real operational problems.