# high entropy · redacted
Secrets detection
Pattern matching and entropy analysis find leaked credentials, API keys and tokens. Reports redact them by default.
Ten disciplines. One engagement.
An end-to-end penetration tester and vulnerability scanner for Mac. Discover, test, validate, map and report from one native app, with scope enforcement and approval gates built into the architecture.
| Finding | Severity | ATT&CK | Proof |
|---|---|---|---|
| SQL injection in patient search | Critical | T1190 | Validated |
| IDOR on records API | High | T1078 | Validated |
| Session cookie missing Secure flag | Medium | T1539 | Validated |
Illustration with fictional data. Interface is in development and will change.
End to end
A real assessment is not one tool run once. It is ten kinds of work that have to agree with each other. Aegis does all ten against the same scope, evidence and findings.
Network scanning with quick, standard, full and custom port profiles, service detection and OS fingerprinting.
Checks for SSRF, XXE, path traversal, open redirects, IDOR, CSRF, cross-site scripting and insecure cookies across websites, REST and GraphQL.
A dedicated SQL injection engine with database fingerprinting, built for the applications that hold the most sensitive records.
An intercepting proxy with TLS termination, plus detection of SSL stripping and certificate pinning.
Validation of cloud configuration and identity controls alongside the application and the network.
Secrets detection, dependency auditing, static analysis, infrastructure-as-code checks and SBOM generation.
Findings are confirmed with controlled proof, so the report lists what is real instead of what a scanner guessed.
Findings are mapped to MITRE ATT&CK and joined into an attack graph that shows how one weakness leads to the next.
An operator that plans and runs the next step for you, and stops for your approval before anything that carries risk.
Evidence vault, client-ready reports, remediation tracking and revalidation to prove a fix actually worked.
Proof over assumptions
Scanner output is a list of maybes. Aegis validates each candidate with a controlled check and stores the request, response and result as evidence.
Attack paths over isolated findings
Three medium findings can add up to one critical route to the data. Aegis maps findings to MITRE ATT&CK techniques and builds the graph, so the report explains the risk the way an attacker would use it.
Remediation follows the same graph: fix the step that breaks the most paths first.
Vulnerability scanning
Point Aegis at a repository and it scans the code, the dependencies and the infrastructure files, all on your Mac.
Pattern matching and entropy analysis find leaked credentials, API keys and tokens. Reports redact them by default.
Reads common dependency manifests and checks them against the OSV vulnerability database. Only the package name and version are sent.
Language-aware patterns for common vulnerability classes, with optional Semgrep or OpenGrep rulesets if you already use them.
Misconfiguration rules for Terraform, Kubernetes YAML and Dockerfiles, with optional Trivy or Grype container image scans.
Generate a CycloneDX software bill of materials and export findings as SARIF for the tools your team already runs.
Generate pipeline gate configurations so a build fails when a finding crosses the threshold you set.
Bounded autonomy
The operator can plan a sequence of checks and carry it out, which saves hours on a large scope. It cannot leave the scope, and it stops for approval before any step that could affect the target.
Target portal.example.test is in scope for this engagement. This step sends test payloads to 6 parameters.
Built for regulated work
Sector-aware defaults tune the testing and the report for the organizations that hold the most sensitive records.
Clinics, labs and health software, where web and API testing protects patient records.
Firms and legal platforms, where client confidentiality depends on access controls that hold.
Financial services and payment systems, where findings have to stand up to an auditor.
One tool, three ways to work
Scan your own repositories for vulnerabilities, leaked secrets and dependency risk. A light workflow with no engagement overhead.
Workspaces and engagements with scope control, authorization, evidence capture, a risk register, retest tracking and audit logging.
Authorized assessments for clients, with formal scope, an evidence workflow and client-ready reports.
In development. Features and requirements may change before release.
Penetration testers, internal security teams and developers who want to test systems they own or have written permission to test. It is a professional tool for authorized work.
No. Testing systems without the owner's permission is illegal in most places. Aegis requires you to define scope and record authorization before active testing, and it blocks targets outside that scope.
No. It does the repetitive work, keeps the evidence straight and shows how findings connect. Judgment about what matters to a particular organization still belongs to a person.
New targets start in passive mode. Anything active needs explicit approval, and a kill switch stops all activity at once. As with any security testing, agree on timing and scope with the system owner first.
No. Scanning runs locally. The only outside lookups are optional vulnerability database queries, which send a package name and version, and the connections you authorize to your own targets.
Aegis is in active development. Email us to follow along or to ask about early access.
Aegis is in development at Forensic Intelligence.
MITRE ATT&CK is a registered trademark of The MITRE Corporation. OWASP and ZAP are trademarks of their respective owners, as are Semgrep, Trivy, Grype, Terraform, Kubernetes, GitHub, GitLab and Azure DevOps. Aegis is not affiliated with any of them.