Artifact Pinning & Hash Gate

1. 01

   Concepts & system design

   60 min· read · map the mechanismtimer

   R37 — SLSA provenance & Sigstore concepts

   R12 — Safetensors

   R39 — Threat modeling provenance

   R13 — Qdrant payload indexing

   Read the assigned sections; map the mechanism, threat model and trade-offs.

2. 02

   Project implementation

   120 min· build · integrate · debugtimer

   Implement revision pinning and SHA-256 hash verification for model/embedding artifacts (D12). *Lab:* Modify a single byte in a local model/config file and prove the startup integrity gate rejects it.

   Build document chunking that stamps every chunk with immutable document_id, chunk_id, sha256_hash, tenant_id, source_uri, and ingestion_version.

   Build, integrate, debug and improve the increment in the active project.

3. 03

   Security lab

   60 min· attack your own systemtimer

   No separate task text for this block on day 17 — the sourced work sits in the block above. Keep the time box.

   Test the abuse case, failure path, authorization bypass or resource limit.

4. 04

   Verification & evidence

   30 min· test · record evidencetimer

   Hash-gate tampering rejection test + chunk provenance metadata unit tests.

   Add or run regression tests; record the 9-point evidence record.

5. 05

   Deliberate practice & Socratic review

   30 min· explain aloud · update gapstimer

   Answer the 8 Socratic questions without notes: What? Why? How? What if? Why not? How can I prove it? Where can it break? What have I learned?

   Re-explain the hardest concept aloud, inspect one edge case, update the gap log.

6. total5 h

• R12Safetensors Documentation↗ huggingface.co

  Zero-copy safe tensor format overview vs. unsafe pickle deserialization risks.

• R13Qdrant Documentation↗ qdrant.tech

  Collections, payload metadata indexing, pre-filtering, multitenancy partitioning, and snapshots.

• R37SLSA & Sigstore↗ slsa.dev

  Supply-chain Levels for Software Artifacts (SLSA), build provenance attestations, and Cosign artifact signing.

• R39OWASP Threat Modeling↗ owasp.org

  Four-question framework & STRIDE: assets, trust boundaries, data flows, threat identification, and mitigation tracking.

Read only the named sections needed for today's task. Reading sits inside the 60-minute concepts block; it is not extra time.

• docsslsa.dev (OpenSSF) · framework overview
  SLSA • Supply-chain Levels for Software Artifacts

  Defines build provenance tracks and levels, the vocabulary you need to state what your artifact hash gate actually proves about how a model or embedding artifact was produced.

  R37moved 2026-11-11
• videoCNCF [Cloud Native Computing Foundation] · 34 min
  Securing the Supply Chain with Sigstore Artifacts Signatures at Scale

  Shows keyless Sigstore signing and verification of artifacts at scale, the operational counterpart to pinning a model revision and checking a SHA-256 digest at startup.

  R37checked 2026-11-11
• videoAdam Shostack · 1 min
  The Four Question Framework For Threat Modeling

  The 60-second statement of what/where/response/review that frames the artifact-pinning gate as a threat-model decision rather than a checksum checkbox.

  R39checked 2026-11-11
• articleChainguard Academy · short procedure
  How to sign blobs and standard files with Cosign

  Concrete cosign sign-blob / verify-blob commands with bundles, the exact primitive you reuse to sign and later verify a model or config artifact byte-for-byte.

  R37checked 2026-11-11
• labSigstore Documentation · hands-on quickstart
  Sigstore Quickstart with Cosign

  Guided signing and verification of a blob you create, which you can adapt into the D12 integrity gate that must reject a single flipped byte.

  R37moved 2026-11-11

A lab is complete only when it produces evidence. Installing PyRIT, Garak, Presidio or TransformerLens is not completion. Saved on this device only.