Fennec · Code security scanner

Security checks fast enough that nobody turns them off.

Fennec scans every change in seconds for security bugs, leaked secrets, vulnerable libraries and quantum-weak encryption. One tool, on every laptop and pull request. Your code never leaves.

39 million

passwords, API keys and tokens detected in code on GitHub in a single yearGitHub, March 2025

The same hours, twice

A cloud key on its way to production

A software company, Thursday afternoon. A developer finishes a payment feature for tomorrow’s release. A cloud access key is still pasted in a config file.

TimeWithout FennecWith Fennec
16:05The developer commits. The security scan took minutes, so the team switched it off last month.The developer commits. Fennec checks the change in seconds and stops the key before it enters history.
16:20The key enters the repository history. The pull request passes a quick review and merges.On the pull request, Fennec shows only what this branch adds: one unchecked input, with the line.
Fri 09:00The release ships with the key and an unchecked input in the new code.The developer fixes the input from the finding. The release ships on time.
Fri 15:00A customer’s security review asks for scan evidence. The team has none to send.The customer’s security review gets a standard SARIF report from the release.
How it endsA live key in history, a bug in production.A clean release, on schedule, with the evidence.

Why Fennec

Six reasons to choose Fennec

01 · Developer speed

A scanner fast enough to stay switched on

When a scan takes minutes, developers disable it and bugs ship with the next release. Fennec scans in seconds, runs where developers already work and shows only the issues each change adds. The check stays on, and problems are fixed at the keyboard, where fixing costs least.

  • Runs on every save, commit and pull request
  • Diff mode shows only what a branch adds
  • Findings inline in VS Code as developers type
  • Automatic fixes for supported issues
  • Terminal review that opens the exact line in your editor

02 · One tool

One tool instead of three separate scanners

Many companies buy code, secret and dependency scanning separately, and still cannot say what is in their code. Fennec checks all of it in one pass, plus configuration files and encryption. One tool to roll out, one report to read.

  • Security bugs in your own code, across major languages
  • Follows untrusted input across files to where it harms
  • Passwords, keys and tokens stopped before they are committed
  • Open-source libraries with known vulnerabilities
  • Risky settings in Docker, Terraform, Nginx and other configs

03 · Quantum readiness

The crypto inventory banks are now asked for

Banks, defense suppliers and critical infrastructure are being asked where their encryption stands. RSA and elliptic-curve crypto must be replaced on CNSA 2.0 timelines, the US schedule for quantum-safe encryption. Fennec finds every use, scores your readiness and writes the inventory in one run.

  • Finds RSA, DSA, elliptic-curve and Diffie-Hellman use
  • Migration deadline shown for every finding
  • Credits quantum-safe algorithms such as ML-KEM and ML-DSA
  • A readiness percentage you can show the board
  • Crypto bill of materials (CBOM) in CycloneDX format

04 · Fits your stack

Switch without rewriting the rules you trust

Your team has written rules for the scanner it uses today. Fennec reuses them without a rewrite and plugs into the places code is reviewed, from pull requests to any CI pipeline. Even AI coding agents can check their own code before you see it.

  • Existing custom rules reused, no rewrite
  • GitHub Action and GitHub App for pull request checks
  • Any CI pipeline, failing builds by severity
  • MCP server lets AI coding agents check their code
  • Findings become tracked cases in AutoSecOps

05 · Accuracy

Fewer false alarms, measured on your own code

False alarms teach developers to ignore a scanner. Every Fennec rule is tested in CI against vulnerable code and look-alike safe code. Known false-positive patterns are published, and in the pilot accuracy is measured again on your code, not ours.

  • Each rule must fire on its vulnerable test case
  • Look-alike safe code must produce zero findings
  • Rules labeled by tier, so teams tune noise
  • Baselines split new, recurring and resolved findings
  • Batch ignore or baseline, previewed before anything changes

06 · Evidence

Audit and customer evidence without extra work

Software companies face customer security reviews and SOC 2 audits. Banks face strict regulators. AI now writes more of the code, and Fennec checks all of it. The same scans produce the standard reports auditors and customers ask for.

  • Standard SARIF reports, including for GitHub code scanning
  • JSON output for any tool or pipeline
  • Audit reports for SOC 2 and customer reviews
  • Crypto inventory for security reviews

Before and after

What changes with Fennec

WhatBeforeWith Fennec
Scan timeMinutes, so developers switch it offSeconds, on every commit
ToolsSeparate code, secret and dependency scannersOne tool, one pass
Leaked keysFound after they reach historyStopped before the commit
Findings per changeThe whole backlog, every timeOnly what the branch adds
Quantum readinessNobody knows where weak crypto isInventory and readiness score in one run

Straight talk

What Fennec will not do

Fix every finding for you

Automatic fixes cover supported issues. Other findings point to the exact line and data flow, and a developer makes the change.

Check dependencies fully offline

Code, secret and crypto scans run fully offline. Dependency checks look up the public OSV vulnerability database.

Promise zero false positives

Some rules are heuristic. Fennec labels every rule by tier, publishes its known false-positive patterns and re-measures accuracy on your code.

How you start

Three repositories, two weeks, a clear report

  1. 1

    Pick three repositories

    Choose three repositories that matter. Read-only access is all we need, and we connect them in an hour.

  2. 2

    First scan, same day

    The first full scan runs the day you connect. Developers install nothing.

  3. 3

    Two weeks of scans

    Every pull request is scanned for two weeks, and we tune the rules with your team.

  4. 4

    Results on the table

    Real findings, live secrets, dependency risk, crypto readiness and scan times, with accuracy measured on your code.

Questions buyers ask

Questions about Fennec

What is Fennec, in one sentence?

Fennec is Singularity’s code security scanner: in seconds it checks your code for security bugs, leaked secrets, vulnerable open-source libraries, risky configuration and quantum-weak encryption, on developer laptops, in pull requests and in any CI pipeline.

Can Fennec run on-premises or in an air‑gapped network?

Yes. Fennec runs on your own machines, on macOS, Linux and Windows, and your code never leaves them. An air-gapped install is available. Code, secret and crypto scans run fully offline; only dependency checks look up the public OSV vulnerability database.

Who decides what blocks a build and what gets ignored?

Your team does. With Fennec you choose which severity fails a build. Accepting a finding takes an in-code note that reviewers see in the pull request, and bulk changes are previewed before anything is written. Automatic fixes apply only when you run them.

Will Fennec slow my developers down?

No. Fennec scans in seconds, fast enough to run before every commit, and shows only the issues a branch adds rather than the whole backlog. Developers see findings where they already work: the command line, VS Code, pre-commit and pull requests.

Won’t it bury my team in false positives?

Fennec tests every rule against vulnerable code and against look-alike safe code, which must produce zero findings. Each rule is labeled by tier so you can tune noise, and in the pilot Fennec’s accuracy is measured again on your own code.

Every feature of FennecThe full list for your evaluation team: 38 capabilities in 6 groups.

What it checks

  • Security bugs in your own code, across the major programming languages
  • Follows untrusted input, even across files, to where it causes harm
  • Knows popular frameworks such as Django, Spring, Rails and Laravel
  • Passwords, API keys and tokens, caught before they reach the repository
  • Open-source dependencies with known vulnerabilities
  • Risky settings in Docker, Terraform, Nginx, Apache, HAProxy, YAML, JSON, XML
  • Custom rules, and existing rules reused without a rewrite

Quantum readiness

  • Finds RSA, DSA, elliptic-curve and Diffie-Hellman use that quantum computers could break
  • Shows the CNSA 2.0 migration deadline for each finding
  • Recognizes quantum-safe algorithms such as ML-KEM, ML-DSA and hybrids
  • Scores your migration with a readiness percentage
  • Crypto bill of materials (CBOM) in CycloneDX format, in one run

Fixing and triage

  • Automatic fixes for supported issues, applied in place
  • Diff mode shows only the problems a branch adds
  • Shared baselines split findings into new, recurring and resolved
  • Interactive terminal review that opens the exact line in your editor
  • Mark, ignore or baseline many findings at once, with a preview
  • Accept a finding with an in-code note reviewers see in pull requests
  • Each rule labeled by tier, so teams can tune noise

Reports and outputs

  • Standard SARIF reports, including for GitHub code scanning
  • JSON output for any tool or pipeline
  • Terminal output during local scans
  • Audit reports for customer security reviews and SOC 2 evidence

Where it runs

  • Fast enough to run on every save, commit and pull request
  • Command line, on every developer laptop
  • VS Code, with findings inline as developers type
  • Pre-commit hook stops secrets and new issues before they are committed
  • GitHub Action and GitHub App for pull request checks
  • Any CI pipeline, failing builds by severity
  • MCP server lets AI coding agents check their own code
  • AutoSecOps, where findings become tracked cases

Deployment and control

  • Install with npx, pipx, cargo or a single binary
  • Runs on macOS, Linux and Windows
  • Your code never leaves your machines
  • Air-gapped install; code, secret and crypto scans run fully offline
  • Dependency checks look up the public OSV vulnerability database
  • Releases verified with published checksums and build attestations
  • Single sign-on for team access
Scans
Code, secrets, dependencies, configuration files and quantum-weak crypto, in one tool
Languages
Major languages, including JavaScript, TypeScript, Python, Go, Java, C#, PHP, Ruby, Swift and Rust
Works with
Command line, VS Code, pre-commit, GitHub pull requests, any CI, AI coding agents, AutoSecOps
Outputs
SARIF, JSON, terminal and CycloneDX crypto bill of materials (CBOM)
Platforms
macOS, Linux and Windows; air-gapped install available
Data
Code never leaves; only dependency checks query the public OSV database

Built for

  • Software and SaaS companies
  • Banks and fintech
  • Defense and public sector
  • Telecom and e-commerce
  • Software agencies
  • AI-assisted teams

Works with

  • Command line
  • VS Code
  • Pre-commit checks
  • Pull requests
  • Docker
  • Terraform
  • Nginx
  • AutoSecOps

Book a demo

Scan three of your repositories in two weeks

Pick three repositories and a date. You get real findings, leaked secrets and a crypto readiness score from your own code.

Architecture, deployment and integration details are shared under NDA.