CompTech AIOps - tracking register
Current issues, milestones, phases, ADRs, and documentation records for the comptech-aiops effort.
This page is the public-safe tracking register for the current work. It names issues, decisions, and artifacts, but does not include secret values.
GitHub Issues
| Repo | Issue | Purpose | Milestone | Phase |
|---|---|---|---|---|
comptech-lab/comptech-aiops | #1 | Establish ADR-driven change control and strict changelog. | Phase 0 - Governance Foundation | phase:0-foundation |
comptech-lab/comptech-aiops | #2 | Register Cloudflare token in GitHub secrets. | Phase 0 - Governance Foundation | phase:0-foundation |
comptech-lab/comptech-aiops | #3 | Inventory ze-tower for platform planning. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #4 | Provision office MinIO access and buckets. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #5 | Store MinIO admin details in approved local custody. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #6 | Record custom portal and direct model serving. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #7 | Record Go/Gin as portal backend framework. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #8 | Record React/Vite/TypeScript frontend stack. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #9 | Clarify ze-tower control-plane boundary. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #10 | Clarify cluster-resident GitOps controllers. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #11 | Record agent runtime architecture. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #12 | Record containerized control plane and network model. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #13 | Record Phase 1 tooling baseline and scaffold monorepo. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #14 | Fix validation workflow npm install path. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #15 | Bootstrap GitHub Actions deployment to ze-tower. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #16 | Configure deployment secrets and run dry-run deploy. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #17 | Install the ze-tower self-hosted runner. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #18 | Run first non-dry-run ze-tower deployment. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #19 | Remove legacy OpenWebUI container from ze-tower. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #20 | Record public blog documentation mirror ADR. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #21 | Define ze-tower backup and restore baseline. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #22 | Implement encrypted ze-tower backups and first restore drill. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #23 | Define Restic retention and prune policy. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #24 | Define office MinIO backup and replication strategy. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #25 | Define real service contracts for the AIOps platform. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #26 | Implement contract-shaped service stubs and tests. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #27 | Add shared service contract helpers. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #28 | Add Postgres-backed operational tables. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #29 | Wire portal API to internal service stubs. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #30 | Implement approval and evidence API behavior. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #31 | Implement orchestrator approval-wait workflow. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #32 | Implement RAG ingestion pipeline skeleton. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #33 | Implement durable RAG metadata and source object storage. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #34 | Build the operator portal MVP. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #35 | Expose the operator portal on the trusted ze-tower LAN ingress path. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #36 | Add deterministic RAG embeddings and Qdrant indexing. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #37 | Add reviewed RAG knowledge pack and retrieval evaluation fixtures. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
comptech-lab/comptech-aiops | #38 | Build the first operator chat console. | Phase 1 - Full Platform Foundation | phase:1-platform-foundation |
zeshaq/comptech-lab | #1 | Add this comptech-aiops docs section to the blog. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #2 | Refresh this public-safe docs mirror from the canonical repo. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #3 | Publish public backup and restore baseline. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #4 | Publish public encrypted backup implementation status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #5 | Publish public Restic retention policy. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #6 | Publish public office MinIO backup and replication strategy. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #7 | Publish public AIOps service contract baseline. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #8 | Publish public contract-shaped service stub implementation. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #9 | Publish public shared service contract helper implementation. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #10 | Publish public Postgres operational store implementation. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #11 | Publish public portal internal client wiring status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #12 | Publish public approval and evidence API status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #13 | Publish public orchestrator approval-wait workflow status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #14 | Publish public RAG ingestion pipeline status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #15 | Publish public durable RAG storage status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #16 | Publish public operator portal MVP status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #17 | Publish public LAN portal exposure status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #18 | Publish public RAG semantic indexing status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #19 | Publish public RAG knowledge pack and retrieval evaluation status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
zeshaq/comptech-lab | #20 | Publish public operator chat console status. | CompTech AIOps Documentation | phase:comptech-aiops-docs |
ADRs
| ADR | Decision |
|---|---|
0001 - GitHub Repo As Operational Memory | Use the comptech-aiops GitHub repo as durable project memory. |
0002 - Local-First RAG On ZE Tower | Superseded historical Open WebUI/Ollama RAG experiment. |
0003 - ADR-Driven Change Control | Require issue, milestone, phase, ADR, and changelog linkage for mutating work. |
0004 - GitHub Secrets As Secret Custody | Use GitHub secrets as the project secret custody system until superseded. |
0005 - Office MinIO For AIOps Object Storage | Use office MinIO as the initial object storage backend. |
0006 - Custom Portal And Direct Model Serving | Use a custom portal and direct vLLM/TGI-class serving; exclude Open WebUI and Ollama. |
0007 - Go Gin For Custom Portal Backend | Use Go/Gin for the portal backend/API. |
0008 - React Vite TypeScript For Custom Portal Frontend | Use React, Vite, TypeScript, TanStack, Tailwind, and live updates. |
0009 - ze-tower Control Plane And Target Infrastructure Boundary | Keep ze-tower as control plane; target infrastructure remains external. |
0010 - Cluster-Resident GitOps Controllers | Keep Argo CD/OpenShift GitOps inside target clusters. |
0011 - Agent Runtime Architecture | Use LangGraph, Temporal, MCP, OPA, LiteLLM, Postgres, Qdrant, and MinIO. |
0012 - Containerized AIOps Control Plane On ze-tower | Run the AIOps control plane as containers on ze-tower. |
0013 - Docker Network Boundary Model | Segment Docker networks by ingress, app, model, observability, and target access. |
0014 - Full Phase 1 Tooling Baseline | Deploy the full Phase 1 control-plane tooling baseline. |
0015 - GitHub Actions Deployment Bootstrap To ze-tower | Use manual GitHub Actions deployment bootstrap to ze-tower. |
0016 - Public Blog Documentation Mirror | Keep this blog section as a public-safe mirror of the canonical repo. |
0017 - ze-tower Control Plane Backup And Restore Baseline | Use encrypted Restic backups in MinIO and require a restore drill before production-changing authority. |
0018 - Restic Retention And Prune Policy | Keep a conservative Phase 1 snapshot set and require explicit apply mode for prune operations. |
0019 - Office MinIO Backup And Replication Strategy | Protect AIOps MinIO buckets by replication or export to an independent target outside the primary MinIO failure domain. |
0020 - AIOps Service Contract Baseline | Define portal, orchestrator, MCP, RAG, evidence, event, and health/readiness contracts. |
0021 - Postgres Operational Store | Use Postgres as the first operational store for tasks, approvals, evidence metadata, and audit records. |
0022 - Orchestrator Approval-Wait Workflow | Link task enqueue, approval request, operator decision, task transition, and orchestrator resume/stop notifications. |
0023 - RAG Ingestion Pipeline Baseline | Establish deterministic source ingestion, chunking, query, and citation behavior before durable RAG storage is wired. |
0024 - Durable RAG Storage Baseline | Store RAG source, job, and chunk metadata in Postgres and reviewed source object copies in MinIO. |
0025 - Operator Portal MVP | Expose safe Phase 1 workflows in the custom portal while keeping the Go/Gin portal API as the only frontend-facing boundary. |
0026 - LAN Operator Portal Ingress | Expose only the operator portal ingress gateway on the trusted ze-tower LAN path while keeping internal services hidden. |
0027 - RAG Embedding And Qdrant Indexing Baseline | Use deterministic local embeddings and Qdrant as the first rebuildable semantic retrieval baseline. |
0028 - Reviewed Knowledge Pack And Retrieval Evaluation Baseline | Track reviewed RAG knowledge packs and retrieval evaluation fixtures in GitHub before broad knowledge ingestion. |
0029 - Operator Chat Console | Add a planning-only portal chat console that creates tracked tasks, queries RAG, and shows proposed plans with citations. |
Changelog Entries
| Date | Issue | Summary |
|---|---|---|
| 2026-05-22 | comptech-aiops#1 | Created strict governance docs, ADR process, issue/PR templates, and changelog. |
| 2026-05-22 | comptech-aiops#2 | Registered CLOUDFLARE_API_TOKEN in GitHub secrets and documented the custody decision. |
| 2026-05-22 | comptech-lab#1 | Created the public docs section for comptech-aiops. |
| 2026-05-22 | comptech-aiops#6-#13 | Recorded the custom portal, frontend/backend stack, runtime architecture, network model, and Phase 1 tooling baseline. |
| 2026-05-22 | comptech-aiops#15-#18 | Bootstrapped GitHub Actions deployment, installed the self-hosted runner, and ran the first real ze-tower deployment. |
| 2026-05-22 | comptech-aiops#19 | Removed the legacy rag-open-webui container while preserving old data paths. |
| 2026-05-22 | comptech-aiops#20 | Added ADR 0016 requiring public blog mirror updates or explicit skip records. |
| 2026-05-22 | comptech-lab#2 | Refreshed this public-safe docs mirror from current canonical records. |
| 2026-05-22 | comptech-aiops#21 | Added the ze-tower backup and restore baseline, including restore drill gate. |
| 2026-05-22 | comptech-lab#3 | Added this public-safe backup and restore baseline page. |
| 2026-05-22 | comptech-aiops#22 | Implemented encrypted Restic backups for ze-tower and completed the first restore drill. |
| 2026-05-22 | comptech-lab#4 | Refreshed this public-safe backup implementation status. |
| 2026-05-22 | comptech-aiops#23 | Defined the Restic retention and prune policy, including dry-run and apply workflow controls. |
| 2026-05-22 | comptech-lab#5 | Refreshed this public-safe retention policy status. |
| 2026-05-22 | comptech-aiops#24 | Defined the office MinIO backup and replication strategy. |
| 2026-05-22 | comptech-lab#6 | Refreshed this public-safe MinIO protection status. |
| 2026-05-22 | comptech-aiops#25 | Defined the AIOps service contract baseline. |
| 2026-05-22 | comptech-lab#7 | Published this public-safe service contract baseline. |
| 2026-05-22 | comptech-aiops#26 | Implemented contract-shaped service stubs and tests, then corrected ingress health routing during deployment verification. |
| 2026-05-22 | comptech-lab#8 | Refreshed this public-safe service stub implementation and deployment verification status. |
| 2026-05-22 | comptech-aiops#27 | Added shared service contract helpers. |
| 2026-05-22 | comptech-lab#9 | Refreshed this public-safe shared helper implementation status. |
| 2026-05-22 | comptech-aiops#28 | Added Postgres-backed operational tables. |
| 2026-05-22 | comptech-lab#10 | Refreshed this public-safe operational store implementation status. |
| 2026-05-22 | comptech-aiops#29 | Wired and deployed portal API calls to internal service stubs. |
| 2026-05-22 | comptech-lab#11 | Refreshed this public-safe internal client wiring and deployment verification status. |
| 2026-05-22 | comptech-aiops#30 | Implemented and deployed approval and evidence API behavior. |
| 2026-05-22 | comptech-lab#12 | Refreshed this public-safe approval and evidence API deployment status. |
| 2026-05-23 | comptech-aiops#31 | Implemented and deployed the first orchestrator approval-wait workflow. |
| 2026-05-23 | comptech-lab#13 | Refreshed this public-safe approval-wait workflow deployment status. |
| 2026-05-23 | comptech-aiops#32 | Implemented and deployed the first RAG ingestion pipeline skeleton. |
| 2026-05-23 | comptech-lab#14 | Refreshed this public-safe RAG ingestion pipeline status. |
| 2026-05-23 | comptech-aiops#33 | Implemented durable RAG metadata and source object storage. |
| 2026-05-23 | comptech-lab#15 | Refreshed this public-safe durable RAG storage status. |
| 2026-05-23 | comptech-aiops#34 | Implemented and deployed the first operator portal MVP. |
| 2026-05-23 | comptech-lab#16 | Refreshed this public-safe operator portal MVP status. |
| 2026-05-23 | comptech-aiops#35 | Exposed the operator portal through a single trusted-LAN ingress path. |
| 2026-05-23 | comptech-lab#17 | Refreshed this public-safe LAN portal exposure status. |
| 2026-05-23 | comptech-aiops#36 | Added deterministic RAG embeddings, Qdrant indexing, vector-first query, and an index rebuild endpoint. |
| 2026-05-23 | comptech-lab#18 | Refreshed this public-safe RAG semantic indexing status. |
| 2026-05-23 | comptech-aiops#37 | Added the first reviewed RAG knowledge pack and retrieval evaluation fixture. |
| 2026-05-23 | comptech-lab#19 | Refreshed this public-safe RAG knowledge pack and retrieval evaluation status. |
| 2026-05-23 | comptech-aiops#38 | Added the first operator chat console with tracked tasks, RAG citations, and planning-only proposed plans. |
| 2026-05-23 | comptech-lab#20 | Refreshed this public-safe operator chat console status. |
Current Deployment Evidence
- Canonical commit
8893e8dimplemented durable RAG storage; public mirror commit96ce79dpublished the public-safe status. - Canonical Validate run
26306911760, blog Deploy run26306911894, and finalze-towerDeploy run26307228961passed. ze-towersmoke verified ans3://RAG source object reference, Postgres-backed job and chunk reads after restartingrag-ingestion, direct RAG query citations, portal/api/v1/knowledge/querycitations, and a final health check on127.0.0.1:8080.- The active
ze-towerGitHub environment MinIO endpoint secret was corrected to the reachable HTTPS API endpoint through GitHub Actions secret custody. No secret values are published here. - Canonical commit
d26ceafimplemented the operator portal MVP. - Canonical Validate run
26308316770andze-towerDeploy run26308349061passed. ze-towersmoke verified the rendered portal, readiness dependencies, a non-secret execution task, approval decision, task event history, portal-mediated RAG citation query, inventory route, evidence metadata lookup, and the standard health script.- Desktop and mobile browser screenshots rendered the portal successfully through an SSH tunnel to the deployed ingress.
- Canonical commit
1008d9cexposed the portal ingress on the trusted LAN path; canonical evidence commit3b5585erecorded the verification. - Public mirror commits
5b4df95and585019epublished this status and evidence. - Canonical Validate runs
26309420527and26309600604,ze-towerDeploy runs26309482441and26309633259, and blog Deploy runs26309420781and26309600523passed. - LAN smoke verified portal health, readiness, the placeholder operator profile, the direct browser render at desktop and mobile viewport sizes, and the remote health script against the configured LAN bind. Credentials and private raw evidence are not published here.
- Canonical commit
e45119eadded deterministic local embeddings, Qdrant indexing, vector-first query with keyword fallback, and an internal index rebuild endpoint. Public mirror commit6d3f019published this public-safe status. - Canonical Validate run
26310550995,ze-towerDeploy run26310590640, and blog Deploy run26310550843passed. - Public-safe smoke verified
rag-ingestionreadiness with Postgres, MinIO, Qdrant, and deterministic embeddings; a reviewed fixture reachedembedding_state: indexed; direct RAG query returned a citation; the index rebuild endpoint reported indexed chunks; portal-mediated knowledge query returned citations; and the standardze-towerhealth script passed. - Canonical commit
0f5b57fadded the first reviewedphase1-control-planeknowledge pack and local retrieval evaluation fixture. Public mirror commit9bf636epublished this public-safe status. - Canonical Validate run
26311183879,ze-towerDeploy run26311226062, and blog Deploy run26311184138passed. - Public-safe smoke verified LAN readiness, confirmed the synced
ze-towerworktree contains the retrieval fixture and evaluator script, and ran the standard remote health script successfully. - Canonical commit
8091c11added the first operator chat console. Public mirror commit3e2947cpublished this public-safe status. - Canonical Validate run
26312411114,ze-towerDeploy run26312451974, and blog Deploy run26312411393passed. - Public-safe smoke verified LAN readiness, a portal chat request that returned a queued tracked task, proposed planning envelope, citations, and the no-target-changing-authority note, a deployed portal browser render, and the standard remote health script.
Open Follow-Ups
- Choose and implement the independent MinIO replication or export target from ADR 0019.
- Replace remaining internal service stub responses with contract-compliant orchestrator and MCP gateway behavior.
- Add target-infrastructure knowledge packs, then compare deterministic local embeddings with a model-backed embedding service after more evaluated packs exist.
- Add evidence list/search API support for the operator portal.
- Persist standalone chat session history after the planning-only chat console is deployed and validated.
- Implement Temporal-backed durable orchestrator runtime behind the approval-wait contract.
- Decide whether to archive or delete preserved legacy
/srv/ragdata. - Keep this docs section updated, reviewed without change, or explicitly skipped whenever a governed change lands.