Skip to content
SecHelixv3 alpha
GitHub
DocsContributeSupportWorkbenchGitHub
Using SecHelix

Full Audit

The complete workflow across a repository: map first, select only applicable hypotheses, hunt in parallel lanes, verify independently, fix root causes, prove regressions, gate the release.

The instruction

A full audit is the widest run: it maps the whole repository, resolves applicability across all 21 families, and ends at a release decision. The instruction below is the one the framework documents.

Agent prompt
Use SecHelix for a complete authorized security audit of this repository.
Start STATIC. If a safe local runtime is available, continue in LOCAL.
Do not skip attack-surface mapping or applicability.
Run specialist review in parallel where useful, independently verify High/Critical candidates,
fix root causes, add regression tests, retest, and produce the final release gate.

Two clauses matter more than the rest. Do not skip mapping or applicability stops the agent from pattern-matching the codebase against a generic vulnerability list. Independently verify stops the agent from grading its own work.

Specialist lanes

Seventeen model-neutral role profiles are defined in the repository. A full audit uses whichever lanes the attack surface justifies; it does not activate all of them by default.

LanePrimary ownership
Surface mapperArchitecture, identities, assets, entrypoints, data flows, trust boundaries
AuthAuthentication, sessions, recovery, MFA, federation
AuthZObject and function authorization, BOLA and BFLA, roles, tenant isolation
Injection and webSQL, command, template, and DOM injection, XSS, CSRF, redirects
SSRF and file parserOutbound requests, uploads, paths, archives, parser boundaries
Business logicWorkflow and state-machine abuse outside payment-specific accounting
Payments and accountingMoney, refunds, payouts, currencies, ledger and reconciliation invariants
Race and idempotencyConcurrency, retries, duplicate work, TOCTOU, outcome-unknown handling
Database, RLS, migrationsDatabase authorization, row policies, constraints, functions, migrations
Browser and extensionClient and server bundles, DOM, origins, extension boundaries
Supply chainDependencies, lockfiles, packages, build inputs, provenance
CI/CD and cloudCI identities, workflows, deployments, cloud policy, runtime configuration
AI, MCP, agentAgent identity, prompts, tools, MCP authorization, untrusted context
Privacy and loggingData minimization, telemetry, redaction, retention
Independent verifierBlind reconstruction and active refutation of candidates
Remediation reviewerCanonical root-cause repair, preservation of accounting truth
Regression and release verifierRegression proof, neighbouring tests, release-readiness evidence

Fast triage variant

When the goal is a quick read rather than a release decision, triage keeps the evidence rules and drops the exhaustive coverage pass.

Agent prompt
Use SecHelix for a security triage of this repository.
Prioritize authentication, authorization, business logic, secrets, injection/SSRF/files,
supply chain, dangerous configuration, and AI/MCP surfaces.
Return only evidence-backed candidates and clearly mark anything unproven.

Wording that works

Asking an agent to find vulnerabilities optimises for finding count, which is the wrong objective. The framework recommends this phrasing instead:

Recommended team wording
Use SecHelix to identify and independently verify security boundary failures
in this authorized repository. Optimize for evidence and low false-positive rate,
not finding count.

That sentence captures the core contract: claims must survive verification before they become findings.

Cost and parallelism

  • Partition by paths, boundaries, or state machines. Do not ask every lane for a repository-wide scan.
  • A mapper may fan out read-only work once the scope and graph are stable.
  • Heavy scanners and full test suites run centrally or under an explicit resource budget.
  • Only one remediation lane owns a file at a time, and no agent mutates the release tree during certification.
  • New evidence can send a candidate back to the owning specialist, but it must not bypass verification.