End-to-End Cross-Project Attack Chain

1. 01

   Concepts & system design

   60 min· read · map the mechanismtimer

   R39 — OWASP Threat Modeling

   R05 — MITRE ATLAS kill chains

   R31 — NIST AI RMF risk treatment

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

2. 02

   Project implementation

   120 min· build · integrate · debugtimer

   Trace a multi-hop attack chain across your architecture: poisoned external input -> model influence -> RAG retrieval / tool call -> delegated OAuth identity -> state mutation -> output sink / audit log.

   Update all 3 project threat models and risk registers. Assign every identified gap an accountable owner, concrete code mitigation, regression test ID, and accepted residual risk.

   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 74 — 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

   Cross-project architecture review package and updated risk registers for P1, P2, and P3.

   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

• R05MITRE ATLAS↗ atlas.mitre.org

  ML attack tactics/techniques: prompt injection, data poisoning, model extraction, and exfiltration.

• R31NIST AI RMF 1.0↗ nist.gov

  Govern, Map, Measure, and Manage functions; risk ownership, measurement, and lifecycle documentation.

• 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.

• docsNIST AIRC · framework section
  AI RMF Core

  The Govern/Map/Measure/Manage subcategory text is what lets you attach an accountable owner and a measurement to each gap in the multi-hop chain instead of listing mitigations without owners.

  R31moved 2026-11-11
• videoPrabh Nair · 1:29:40
  Practical Threat Modeling Master Class: STRIDE-Powered

  Builds a live e-commerce threat model from DFD through trust boundaries to a risk register with Jira tickets — the exact artefact refresh you are doing across three projects.

  R39checked 2026-11-11
• videoAdam Shostack
  Data Flow diagrams in Threat Modeling

  Explains the five DFD elements (external entity, process, data store, data flow, trust boundary) so the hop from retrieved document to tool call to OAuth identity is drawn on the diagram where reviewers can see it.

  R39checked 2026-11-11
• articleCompile N Run · guide
  Threat Modeling with STRIDE

  Applies the four-question framework and STRIDE-per-element to each DFD element, and shows keeping the register as code with verified_by links to regression tests — matching the regression-test-ID requirement.

  R39R31moved 2026-11-11
• repoOWASP (GitHub) · desktop + web app
  OWASP Threat Dragon

  Runs locally and stores threat models as JSON you can commit, so the three updated models and their per-threat mitigations live in version control alongside the code they describe.

  R39checked 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.