Every AiGovOps Foundation property, its part in the practitioner ladder, the gate(s) it serves, the
movement it belongs to, and what changes if PRD v3 is approved. Read alongside
PRD-OPEN-SOURCE-UNIFIED.md §2. Sources: the local clones under
~/Downloads/_aigov as of 11 September 2026 and the Library’s plan/strategy documents.
The rule: a property that cannot name its level and its gate is folded into one that can, or retired, in the copy sweep. Nothing on this list is new infrastructure.
| Property | Repo / surface | What it is today | Level(s) | Gate(s) served | Movement | What changes under v3 |
|---|---|---|---|---|---|---|
| Foundation site — the porch | aigovops-foundation/aigovops-foundation-site-redesign-June2026-ken-and-bob · aigovops-foundation.com |
The only front door. One message, four services, one loop. Gate Check in all 22 footers and every hero. Three movements as the IA. EN + ES | 100 (the page), all (the ladder) | Pre-pend (the Gate Check page) | Get to Yes | Gate Check page gains rule box (plain words + YAML), action box, four lens tabs, download, share; Community page becomes the ladder + the four gates + the readiness promise; chatbot intents learn “what level am I” |
| AiGovOps Library (private, members) | aigovops-library · Fly core + CDN hub |
100 verified AI-harm cases (batch 2026-W24, 67 primary / 33 strong-secondary), each with the plain-words gate that would have caught it; #FridayFailFest daily rotation; the 10-question practitioner test; the 1/0/? gate law; runtime primitives T0 SecretsProvider, T1 gate↔secrets, T3 sandbox, T5 capability dial; T10 JCS canonicalizer next; the ledger | all (scenarios), 100 (test), 300 (broker), all (ledger) | Operate (broker); Audit (corpus verdicts) | Recover to Yes | The corpus becomes the exam bank and the scenario bank; the gate sentence becomes the plain-words path’s input; T10 ships first; the ledger mints practice.complete and practice.mark receipts |
| Beacon | aigovops-foundation/aigovops-beacon · GitHub Pages demo · Node server + Studio |
OVERT 1.0-conformant runtime signer: Ed25519, JCS, append-only log, Merkle anchoring, auditor bundle with VERIFY.md. Browser demo generates a key and a real bundle client-side. Profile aigovops-beacon.v1 registration pending Glacis sign-off since May |
100 (page), 300 (broker/holds), 400 (verify) | Pre-pend; Operate; Audit | Get / Stay / Recover | Gains beacon gate check, gate init, hold, coverage, verify --offline, share; Python SDK (v1 F2); agent identity binding (v3 F15); README AAL/IAP line corrected; profile registered |
| Umbrella-GovOps | aigovops-foundation/umbrella-govops · umbrella-conformance CLI (install from source; PyPI not yet published) |
The program layer: laws compiled to versioned UCIDs, crosswalks (NIST AI RMF, EU AI Act, ISO 42001, OECD…), EvidenceBundle binding Beacon receipts to controls, the 41-item policy-as-code vendor checklist |
200 (catalog, CI), 400 (bundle export) | Policy gates; Audit | Get to Yes; Recover | Gate vocabulary (tiers, decisions, constraints, stop conditions) added to the YAML; the community policy catalog with steward review; policy change emits a receipt; PyPI release so the 200 README’s first line works |
| Lantern | aigovops-foundation/aigovops-lantern · Python CLI v0.1.1 |
read, diff, explain; four role lenses; “Beacon signs. Lantern reads.” Open issues: v0.2 web viewer, v0.3 GitHub Action |
all (read), 300+ (verify page), 200+ (Action) | Audit (viewer); Policy gates (Action) | Recover; Get | Renders gate decisions; diff across two gate runs; v0.2 becomes the public verify page (the August plan’s “make Lantern real by making it the verify page”); v0.3 becomes the 200 policy gate in CI |
| Omni / Jeeves | aigovops-foundation/Omni-Rapp-June-2026 · droplet |
The agent estate: 45 agents, crew roles, the Design agent (design mark), twins and envelopes (inert), jeeves numbers, cohort call, letter assembly, story@ intake, the ledger |
all (return) | — (machinery) | the loop | Counts loops and marks by level; Thursday run-of-show gains the five-minute receipt slot; host rota as data; ledger schema for practice.* receipts |
| NCW AI Camp | aigovops-foundation/aigovops-ncw-ai-camp · GitHub Pages |
Cool Tools, Cool Schools, The Rules — 35 mini-camps, 16 tool pages, the Hint-Logs “keep the receipts” method, the 7-point NCW AI Partnership Pledge (EN/ES). Built for AI Expo, 11 Aug 2026 | pre-100 | — | Get to Yes | Becomes the named zero-tech on-ramp: “if you have never opened a terminal, start here”; Hint-Logs is the receipts habit that makes the 100 page familiar; one mini-camp links straight to the Gate Check page |
| Vendor RFI | aigovops-foundation/aigov-ops-open-source-vendor-rfi-rapp-johnston-june-2026 |
Evidence-oriented vendor stack composer (Beacon lens, Umbrella lens); auditor value model; neutral sponsorship framework; the audit-training tool with a scored knowledge check, self-audit and receipt export | 300, 400 | Audit; Policy gates | Recover; Get | The 300/400 procurement activity: score a vendor stack in evidence terms; the self-audit’s receipt joins the member’s chain |
aigovops — Open Source v4 |
aigovops-foundation/aigovops |
Self-hostable governance stack from a clone: zero-dependency Node gate (up --tier 1 = just the gate), Caddy/Keycloak/OpenSearch/Prometheus at tier 4; 193 tests; npm packages and get.aigovops.org planned, not live |
200, 300 | Pre-pend; Operate | Get; Stay | The tier-1 gate is the “gate one real call today” path in the 200 worksheet until the Beacon Python SDK lands; npm publish is the August plan’s wedge |
| aigovops-Replay | aigovops-foundation/aigovops-Replay |
Cryptographically signed receipt system for AI interactions | — | Audit | Recover | Fourth receipt implementation in the estate (the August plan counted four); folds onto the one OVERT envelope or is archived in the copy sweep |
| aigovops-prompt-studio | aigovops-foundation/aigovops-prompt-studio |
Prompt engineering platform with immutable audit logging and version control | — | Policy gates | Get | Its audit log adopts the receipt; otherwise out of the ladder |
| Global Inclusion Matchmaker | (repo named in v1 §10) | The G20 rural and Indigenous resource app | 200 | Pre-pend; Policy gates | Get | The retrofit worked example — reference implementation 2 and the 200 README are one artifact |
| Newsletter / Tuesday Letter | Substack + LinkedIn newsletter; Omni letter_assemble |
Weekly; one Publish click for Ken | all | — | the loop | One line from jeeves numbers: loops closed and marks minted this week, by level, no names; the readiness countdown to 2 December 2027 |
| Glean-Mastery | Glean-Mastery (in _aigov) |
30-minutes-a-day curriculum, levels 200→500, content as data gated by policy-as-code, quizzes as JSON | — | — | — | Confirm scope before reuse (employer IP carve-out). If Foundation-scoped: its content/weeks/NN/day-NN.mdx + data/quizzes format is the template for level kits’ daily practice; otherwise reuse only the idea |
| quantum-stack-intelligence | quantum-stack-intelligence |
(out of scope for this map) | — | — | — | Not part of the ladder; not touched |
Where the corpus cases live
Batch 2026-W24, aigovops-library/docs/data/verified-harms-2026-W24.json, schema 2.0.0, 100 cases,
verdicts 85 × 0 · 13 × ? · 2 × 1. Domains by count: chatbots-llm 17, biometrics-face 17,
content-data-provenance 14, public-benefits 11, healthcare 9, deepfakes-fraud 8, autonomous-vehicles 6,
hiring 5, finance 4, content-moderation 3, education 2, justice 2, safety-physical 2. Each case carries
gate (the plain-words rule), verdict, frameworks, source[], verification. The gate field is
the plain-words path’s input at every level.
Ken’s themes and where the ladder answers them
| Theme (Foundation blog) | Where it lands |
|---|---|
| Agentic Hell — recursive loops, no observability, no provenance | The chain, gap detection, coverage; 300 owns it |
| No More Anonymous Ghosts — agent identity and authority | agent_id + authority bound by the broker (F15); an unbound action is a stop |
| EU AI Act ready by August 2026? | The Omnibus moved Annex III to 2 Dec 2027; the readiness promise names that date |
| Healthcare AI governance, spring 2026 | 300 is built on nH Predict and Cigna PXDX: a model may inform, never replace clinical judgment; a human actually reviews |
| NIST AI RMF vs the real world | UCID crosswalks; policy gates in CI |
| Flow-to-Trust Loop — ten ops disciplines | The four gates are where GovOps hands off to the other nine |