Out-of-pattern changes
A new branch, unsigned commit, or divergent artifact can change release trust.
Supply Chain Sentinel
Decide before risk enters the chain.
SΛDB correlates code, pipeline, dependency, credential, and traffic signals to produce an explainable assessment before an artifact is installed, published, or released.
01Branch trustProtection, signature, creation, and actor identity.
02Pipeline integrityWorkflows, permissions, and publishing.
03Package behaviorScripts, downloads, and artifact divergence.
04Credential exposureSecrets, tokens, and sensitive configuration.
The problem
A trusted chain can be compromised by a single change.
Dependencies, CI/CD automation, and privileged tools expand the surface between code and production. Isolated alerts cannot determine on their own whether an operation should proceed.
Privileged automation
Modified workflows, sensitive permissions, and unexpected publishing must be evaluated together.
Installation-time behavior
Access to secrets, obfuscated scripts, downloads, and external destinations may indicate compromise.
Specialized guards
Six perspectives combined into one decision.
The engine aggregates complementary signals, removes duplicate rationales, and classifies risk and the recommended response.
Branch Trust Guardian
Evaluates branch creation and protection, commit signatures, and actor behavior.
Pipeline Integrity Guard
Analyzes changes to workflows, permissions, OIDC, publishing, and external downloads.
Package Behavior Scanner
Observes install scripts, obfuscated code, precompiled artifacts, and version inconsistencies.
Credential Shield
Identifies access to paths and variables associated with tokens, keys, and credentials.
AI Tool Secret Guard
Considers access to sensitive configuration from AI-assisted development tools.
Exfiltration Guard
Correlates destinations, traffic during CI or installation, and repository-creation anomalies.
Decision flow
From signal to governed response.
The assessment does not reduce the chain to a number: it preserves the reasons, recommended actions, and associated evidence.
Normalize
Organizes repository, pipeline, package, and environment signals.
Assess
Runs the guards applicable to the observed context.
Classify
Classifies risk as LOW, MEDIUM, HIGH, or CRITICAL.
Recommend
Produces a proportional response and its rationale.
Preserve
Generates bound evidence for high or critical risks.
The public page explains the capability. Repository data, assessment forms, scores, hashes, history, and operational controls belong exclusively to the private platform.
Possible responses
Actions proportional to observed risk.
Responses are selected according to the signals and classification while respecting the architecture's authorization boundary.
Monitor or require approval
Intermediate risks may require supervision, sandboxing, or approval before proceeding.
Block installation, pipeline, or release
High-risk signals may produce an explicit recommendation to stop the affected stage.
Isolate, revoke, and preserve
Critical scenarios may combine isolation, token revocation, evidence preservation, and incident creation.
Technical demonstration
Evaluate how Decision Architecture can protect your software supply chain.
Present your development flow and explore a bounded use case with SΛDB.