Skip to content

How the Secure SDLC Evidence Collector compares to adjacent tools

This page is an honest crosswalk: where the collector overlaps with existing tools, where it differs, and where it is the wrong choice. Items marked "honesty check" call out where another tool is better at a specific job. Every claim here can be verified against the linked project's docs.

Positioning in one sentence

The Secure SDLC Evidence Collector is the release-readiness evidence layer: it ingests the output of every scanner and attestation tool, normalises it into a single canonical EvidenceBundle, evaluates it against a control catalog (NIST SSDF 1.1 / 1.2 / AI Profile / FedRAMP 20x KSI / OSPS Baseline), and emits an audit-ready JSON + report. It does not scan, sign, or enforce; it makes the "did this release meet our Secure SDLC bar?" question answerable from one document.

Side-by-side matrix

Product Primary focus Overlap with the collector What's clearly better elsewhere
Chainguard Enforce (docs) Image hardening + admission control for Kubernetes Cosign + attestation consumption Image-level hardening, OCI ecosystem, admission policy at deploy time
Scribe Security (Trust Hub) (docs) End-to-end SBOM + provenance + policy-as-code SaaS SBOM ingest, attestation, evidence model Hosted UI, customer support, signed analytics; collector ships OSS only
GUAC (OpenSSF Incubating, docs) Graph of supply-chain metadata The collector emits a GUAC-ingest container (T6.9) Graph queries, blast-radius analysis, cross-org search
Kusari Trustify (docs) Backend search for supply-chain metadata (contributed into GUAC) None — collector is upstream of Trustify Hosted backend for SBOM search; collector produces input documents
Lineaje (site) SaaS supply-chain integrity + remediation SBOM ingest, dependency risk Vendor-driven remediation, commercial support
OpenSSF Scorecard (docs) Health score for OSS repos Both reference SSDF practices Heuristic scoring on git history; the collector consumes Scorecard as evidence
Snyk / Mend / Sonatype Commercial SCA + SBOM platforms SCA findings via SARIF / OSV Curated vuln intelligence, fixes, dashboard; collector consumes their SARIF
CodeQL reachability / Endor Labs / Semgrep Pro Reachability-aware vuln triage Collector records their verdict via Reachability field (§3.2) Re-deriving reachability — collector never does this itself

Where the collector is the right answer

  • You need a single signed JSON that proves a release followed Secure SDLC, ingestible by GUAC, Kyverno, OPA, Sigstore policy-controller, or a custom verifier.
  • You ship to a regulator (CRA 2026, FedRAMP 20x) that wants a machine-readable evidence pack instead of a PDF.
  • You want SBOM + SARIF + attestation + waiver + AI evals + EPSS + KEV traveling together with the same release_id / commit_sha, with a deterministic structural hash.
  • You distrust scoring theater and want a verdict that calls out the specific missing required evidence (not "92 %").

Where the collector is the wrong answer

  • Honesty check. If you need a hosted UI with trend analysis across thousands of releases, Scribe Trust Hub is built for that. The collector is CLI-first and stateless by design.
  • Honesty check. If you need image-runtime enforcement, Chainguard Enforce has the admission-controller side covered; the collector emits the attestation, not the gate at the runtime edge.
  • Honesty check. If you need a query language over an organisation-wide supply-chain graph, GUAC is the right place — the collector is one of GUAC's many sources (T6.9 closes this with the GUAC adapter).
  • Honesty check. If you need to generate findings, you need a scanner. The collector intentionally does not scan; it consumes scanner output.

What "OSS canonical" means here

  • Apache-2.0, no telemetry, no signed binary download outside GitHub Releases.
  • Every format the collector emits is an open standard: SARIF, CycloneDX (1.5–1.7), SPDX, OpenVEX 0.2.0, OSCAL 1.1.2, in-toto Statement v1 + DSSE, SLSA Provenance v1, OSV Schema, GUAC collector container.
  • The control catalog is YAML; teams override or extend it without forking the project. Five catalogs ship today: NIST SSDF 1.1 (default), SSDF 1.2 preview, FedRAMP 20x KSI, OSPS Baseline, and the AI Profile.

When to combine

Goal Stack
Public OSS release with audit trail Collector + Sigstore (cosign keyless) + SLSA generic generator
Continuous ATO with continuous evidence Collector + GUAC (via T6.9 adapter)
AI / LLM product with SDLC AI Profile Collector + garak + lm-eval + Hugging Face model cards
EU CRA disclosure pipeline Collector with --profile cra-2026 + ENISA intake
FedRAMP 20x evidence pack Collector with --catalog catalog-fedramp-20x-ksi.yaml + OSCAL export

Asking the inverse question

The collector does not aim to replace any of the tools above. The inverse question — "could any of them replace the collector?" — has a single-line answer today: none of them are OSS, format-neutral, and exclusively focused on release-readiness evidence. That niche is where this project lives. Where the niche overlaps with another tool's roadmap, the right move is to interop (GUAC ingest, OpenVEX emission, OSCAL export), not to compete.

If you find this comparison page misrepresents your project, please open a Discussion or PR. The project commits to keeping this page honest — it is part of the public contract.