Skip to content

Progress Tracker

Status of the repository build. "Complete" means written, checked by the automated validators, and with no placeholders. It does not mean every sentence has had expert human review; see Known gaps.

Last updated: 2026-10-24 · Release: v1.1.0

Summary

Area Status Size Verified by
Certification registry Complete 3 supported tracks, 1 planned Official v1.0 exam guides (2026-10-24)
Learning modules 00–15 Complete 16 modules, 139 pages, ~262k words Link check, exam-claims check, snippet compile/run by authors
AI security track (module 12) Complete 16 pages, OWASP LLM Top 10 2025 and agentic mapping Sources checked 2026-10-24
CCAO-F track Complete Overview, 7 domain guides, exercises, scenarios, revision, readiness Coverage matrix: 30/30 objectives
CCAR-F track Complete Overview, 6 exam scenarios, 5 domain guides, exercises, revision, readiness Coverage matrix: 30/30 objectives
CCAR-P track Complete Overview, 7 domain guides, exercises, scenarios, revision, readiness Coverage matrix: 38/38 objectives
Practice questions Complete 135 CCAO-F + 131 CCAR-F + 144 CCAR-P = 410 Schema, keys, distractors, near-duplicates, answer-cueing limits
Mock exams Complete Terminal + browser engines, 9 fixed forms (CCAR-F forms use 4 of the 6 scenarios) Unit tests, scripted CLI run, in-browser test
Hands-on labs Complete 20 labs All 20 pass in mock mode (tools/run_lab_smoke.py)
Real-world projects Complete 5 projects 194 offline tests passing
Learning roadmaps Complete 7, 15, 30, 60 and 90-day plans Link check
Flashcards Complete 106 CCAO-F + 133 CCAR-F + 147 CCAR-P + 94 fundamentals = 480 tools/flashcards.py
Interview prep, cheat sheets, troubleshooting Complete 22 files Link check
Glossary and FAQ Complete 167 glossary terms Link check
Governance files, templates, CI Complete 3 workflows, 4 issue templates, PR template Local equivalents of every CI job pass
GitHub Pages website Complete, deployed MkDocs + Material, 251 pages, search, interactive practice exam Strict build with 0 warnings; 4,354 internal links checked; browser-tested

Quality gates (local run, 2026-10-24)

Check Result
tools/validate_all.py --labs: questions, flashcards, internal links, exam claims, secrets, coverage, forms, web bundle, labs Pass
pytest (repository tests) Pass
Project test suites (5) Pass (194 tests)
markdownlint-cli2 0 errors
External links (tools/check_external_links.py, 264 URLs) All reachable; a few sites block bots (403) and need manual checks
Bandit SAST (Python) No medium/high issues after review; low-severity findings are lab assert self-checks, non-crypto random, and fixed-argument subprocess
Originality scan (2026-10-24) Compared every 10-word sequence against 256 cited sources (250 web pages, the 4 official exam guides, the exam policy and NIST AI 600-1). 18 passages that echoed source wording were rewritten, along with one workflow adapted from the docs. The remaining matches are citations, required code/API syntax, official names and figures, and the official exam blueprint tables.
Code and secret scanning Bandit (Python SAST) reviewed locally; gitleaks, the repository secret scanner, and GitHub secret scanning with push protection run on every push. Snyk is intentionally not used for this personal project.

Independent review (2026-10-24)

Five independent reviews (API accuracy, product facts, question quality, learner experience, code and security) were run on the whole repository. Every critical, high, medium and low finding was fixed or explicitly decided, in four verified commits. Decided, not changed: code repeated across the five projects stays, so each project can be copied out and run on its own.

Known gaps

These are honest limitations of v1.0. None is hidden as "complete".

  1. Live-mode code was not run against the paid API. Every lab and project runs offline in mock mode. Live paths follow the current API rules, and labs 03 and 07 were checked against the real anthropic 2.3.0 SDK using a local stub server, but no request was sent to Anthropic. Expect small fixes on the first live run.
  2. Exam guides change on 2026-11-17. The registry reflects v1.0. Re-verify and update after that date (see the ROADMAP).
  3. Facts labelled Unverified. Search the repository for Unverified: to list them. Most concern plan-specific features, regional or compliance attestations (FedRAMP, HIPAA eligibility per offering), and a few beta details. Each needs confirming against an official page or with Anthropic.
  4. Planned track: CCDV-F has verified facts in the registry but no track content yet.
  5. Expert review: content was written to strict standards and machine-checked, but a human subject-matter review of each module (especially practice-question answer keys) is still recommended. Independent AI reviews of all three banks' answer keys ran on 2026-10-24 and their fixes are applied.
  6. External links: some pages (claude.ai, hhs.gov, iso.org) block automated checks and were not machine-verified.

Next steps

  • Upload images/social-preview.png in GitHub: Settings → General → Social preview

  • Maintainer review of README, disclaimers and licence choice

  • Spot-review a sample of practice questions per domain
  • Optional: run a few labs in live mode to confirm live paths
  • Set up a personal git identity, commit, then push to the personal GitHub repository
  • Turn on GitHub Pages: Settings > Pages > Source: GitHub Actions (the site deploys on the next push to main)
  • After 2026-11-17: refresh the registry from the updated exam guides