Mr.PlanB Logo
    Infrastructure Automation Ledger 2026
    Ledger
    Infrastructure Automation Ledger 2026

    Ansible: Infrastructure Automation Ledger 2026 Scorecard

    The open-source engine inside Ansible Automation Platform: agentless and broad through its collections, with no state file, control plane or AI of its own.

    Edition October 16, 20268 min read

    Where Ansible put its effort, 2016 to 2026

    Each square is one capability area in one year. The brighter it is, the more the area gained in the Infrastructure Automation Ledger that year; 1, 2 and 3 mark that year's biggest pushes.

    ’16’17’18’19’20’21’22’23’24’25’26
    Provisioning breadth1111
    Configuration management depth21
    State, planning and safety
    Language and authoring31
    Orchestration and self-service21
    Scale and control plane
    Security and compliance22
    Bare metal and day-2 operations2
    Ecosystem and integrations1
    AI assistance
    no change+0.5+1+1.5 or moreScores out of 5 per area.

    Biggest pushes overall: Provisioning breadth (+2), Configuration management depth (+1), Language and authoring (+1). Already strong in 2015: Ecosystem and integrations, Configuration management depth, Language and authoring.

    Top focus each year: 2016 Provisioning breadth; 2019 Provisioning breadth; 2020 Provisioning breadth; 2021 Ecosystem and integrations; 2023 Orchestration and self-service; 2024 Configuration management depth; 2025 Language and authoring; 2026 Provisioning breadth. 3 of 11 years show no gain in any area.

    Part of the Infrastructure Automation Ledger (edition 2026-10-16). How the scores work: methodology.

    Snapshot

    • Type: engine (open source; packaged commercially as Red Hat Ansible Automation Platform)
    • Owner: Red Hat, Inc. (IBM subsidiary) employs the core team; developed in the open with a community, no foundation
    • Licence: GPL-3.0-or-later; no licence change found; free with no limits
    • Latest release: ansible-core 2.21.5 (2026-10-05); community package 14.5.0 (2026-10-06); 2.22.0rc1 tagged 2026-10-13
    • Support lifecycle: Three ansible-core versions maintained at once, a new major about every six months in May and November, about 18 to 24 months from GA to end of life; no LTS upstream
    • Next milestone: ansible-core 2.22 GA in November 2026 (published schedule); control node Python 3.13 to 3.15, PowerShell 7 targets
    • Targets: Linux, Windows (SSH officially since 2.18) and network devices through connection plugins; cloud collections for AWS, Azure and Google; vSphere: vmware.vmware 2.12.0 (Red Hat certified, 2026-10-12) and community.vmware 7.0.0 (community, 2026-09-27); Proxmox VE: community.proxmox 2.1.0 (community team, 2026-10-06, created 2025-05-12); Nutanix: nutanix.ncp 2.6.0 (Nutanix, certified, 2026-09-23); Hyper-V: microsoft.hyperv 1.1.0 (certified, 2026-05-19, first release 2026-03-26); XCP-ng: no dedicated collection, only XenServer modules in community.general tested up to XCP-ng 8.0; KubeVirt: kubevirt.core 2.3.0 (kubevirt project, 2026-06-10)
    • Tier: Open-source engine
    • Archetype: Dependable engine on a published calendar
    • Evidence grade: B. Release dates, support windows and collection releases were read from docs.ansible.com, PyPI and GitHub release feeds. Ownership dates come from secondary sources, and the capability steps before 2023 are inferred from release dates because those release notes were not read.

    Scope of the scores: ansible-core is scored together with the collections that reach its targets, because since 2.10 almost every module lives in a collection. Scheduling, a web UI, RBAC and AI belong to the products built on it and are scored under Ansible Automation Platform.

    Scorecard

    DimensionScoreDirection
    Capability today58.6 / 100higher is better
    Momentum since end of 2022+6.7 points (2015 to 2022: +10.1)higher is better
    Credibility19.5 / 25higher is better
    Commercial risknot scored (engine: nothing to buy)
    Technical lock-in4.5 / 25lower is better
    Integrator fit11.5 / 20higher is better
    AI leverage1.0 / 20higher is better
    Ledger Index60.4 / 100balanced weights; see the methodology

    Capability by domain, 2015 to today

    Scores 0 to 5 against fixed anchors; total is weighted to 100.

    Domain (weight)201520192022Today
    Provisioning breadth (15)233.54
    Configuration management depth (12)333.54
    State, planning and safety (12)2222
    Language and authoring (10)333.54
    Orchestration and self-service (10)11.51.52
    Scale and control plane (8)2222
    Security and compliance (10)1.51.51.52.5
    Bare metal and day-2 operations (8)2.5333
    Ecosystem and integrations (8)3.53.54.54.5
    AI assistance (7)0000
    Total / 10041.846.651.958.6

    Year by year: 2015: 41.8 · 2016: 44.3 · 2017: 44.3 · 2018: 44.3 · 2019: 46.6 · 2020: 51.1 · 2021: 51.9 · 2022: 51.9 · 2023: 53.9 · 2024: 55.1 · 2025: 57.1 · 2026: 58.6

    What moved the score

    YearDomainChangePointsTrigger
    2016Orchestration and self-service1 → 1.5+1.02.0: blocks and strategy plugins
    2016Provisioning breadth2 → 2.5+1.52.0: general availability with a larger cloud and network module set
    2019Bare metal and day-2 operations2.5 → 3+0.82.9: Redfish BMC and reboot modules in core before the split
    2019Provisioning breadth2.5 → 3+1.52.9: cloud, vSphere, oVirt and network vendor modules in core before the collections split
    2020Configuration management depth3 → 3.5+1.22.10: Windows and network platform content maintained as collections
    2020Ecosystem and integrations3.5 → 4+0.82.10: collections distribution through Galaxy opens the content ecosystem
    2020Language and authoring3 → 3.5+1.02.10: namespaced, versioned collections on Galaxy
    2020Provisioning breadth3 → 3.5+1.52.10: collections model; vendors maintain their own content on Galaxy
    2021Ecosystem and integrations4 → 4.5+0.8Community package 3.0 (2021-02-18): curated collection bundle on top of ansible-core
    2023Orchestration and self-service1.5 → 2+1.0ansible-rulebook: event-driven runs reach GA with Automation Platform 2.4 (2023-06-27)
    2023Security and compliance1.5 → 2+1.02.16: conditionals with embedded templates on untrusted data fail as unsafe (CVE-2023-5764)
    2024Configuration management depth3.5 → 4+1.22.18: Windows hosts officially managed over the ssh connection plugin
    2025Language and authoring3.5 → 4+1.02.19: data tagging names file and line in errors; native Jinja mode only; language server in the VS Code extension
    2025Security and compliance2 → 2.5+1.02.19: template trust model inverted, only strings from trusted sources render as templates
    2026Provisioning breadth3.5 → 4+1.5microsoft.hyperv 1.0.0 certified (2026-03-26); vmware.vmware certified releases; community.proxmox since 2025-05; still no XCP-ng collection

    Direction, 2023 to 2026

    Each release item carries one theme and one driver (V vision, C customer need, M market window, U upstream, P portfolio).

    DateItemThemeDriver
    2023-06ansible-rulebook: event-driven runs reach GA with Automation Platform 2.4OrchestrationV
    2023-112.16: hardened conditional templating after CVE-2023-5764SecurityC
    2024-112.18: Windows hosts managed over SSH; control node needs Python 3.11 or newerConfiguration managementC
    2025-05community.proxmox collection published on GalaxyProvisioning breadthM
    2025-072.19: data tagging, native Jinja mode only, stricter conditionals; GA in July instead of MayLanguageC
    2025-072.19: template trust model invertedSecurityC
    2025-112.20: control node Python 3.12 or newer, Galaxy v2 API removed, deprecations scheduled for 2.24PackagingU
    2026-03microsoft.hyperv 1.0.0, Red Hat certified Hyper-V collectionProvisioning breadthM
    2026-052.21 on the published date; porting guide lists no notable engine changesPackagingU
    2026-06kubevirt.core 2.3.0 from the kubevirt projectProvisioning breadthU
    2026-09community.vmware 7.0.0 requires vcf-sdk 9.0 or newerProvisioning breadthU
    2026-09nutanix.ncp 2.6.0; legacy modules to be deprecated in a Q4 2026 releaseProvisioning breadthC
    2026-10vmware.vmware 2.12.0, Red Hat certified vSphere collectionProvisioning breadthP
    2026-102.22.0rc1; GA scheduled for November 2026 with PowerShell 7 targetsPackagingU
    • Centre of gravity: Provisioning breadth (6), Packaging (3), Security (2)
    • Blind spots (no items): State and planning, Scale, Bare metal, Ecosystem, AI, Licensing
    • Driver mix: Vision 7%, Customer need 36%, Market window 14%, Upstream 36%, Portfolio 7%
    • Idea-to-need index: 0.13 (0 = demand-pulled, 1 = thesis-pushed)

    Credibility: 19.5 / 25

    CriterionScoreBasis
    Cadence4.5Published May and November rhythm, kept for 2.20 (2025-11-03) and 2.21 (2026-05-18); 2.19 landed 2025-07-21
    Roadmap transparency4.5Public per-release roadmap pages with dates, open tracker, porting guides; feature plans beyond deprecations are thin
    Say-do4Dated releases held; deprecation promises for 2.23 and 2.24 still pending
    Velocity3.52.19 brought several substantive changes; 2.20 and 2.21 were maintenance and platform floors
    Lifecycle stability3About 18 to 24 months per version, no LTS, Python floor rises each cycle, templating breaks in 2.19

    Commercial risk: not scored

    This is an engine: nobody sells it, so there are no commercial terms to score.

    Technical lock-in: 4.5 / 25 (lower is better)

    CriterionScoreBasis
    Formats1Open YAML and plain-text inventory, Ansible specific
    Export path2No state to export; leaving means rewriting playbooks in another tool
    Stack coupling0.5Agentless; needs Python on Linux targets and PowerShell 5.1 or later on Windows
    Hardware / cloud coupling0Any host, fully offline
    Skills and tooling coupling1Widely held skills; open tooling

    Integrator fit: 11.5 / 20

    CriterionScoreBasis
    Partner programme openness3No programme for the engine; anyone can build on it; certified content runs through Red Hat Partner Connect
    Multi-tenancy and self-service0.5None in the engine
    API and infrastructure-as-code3.5CLI, ansible-runner library and ansible-rulebook; no REST API of its own
    Skills and certification4.5Exam-based Red Hat certification (EX294) and a very large talent pool

    AI leverage: 1.0 / 20

    How the product uses AI, scored per year like capability. Only GA counts.

    Criterion20192022TodayLatest step
    AI code and plan generation0002015: None
    AI platform stack0002015: None
    AI-assisted operations0002015: None
    Agent openness1112015: CLI and the ansible-runner Python API; no REST API and no official MCP server for the engine

    Year by year: 2015: 1 · 2016: 1 · 2017: 1 · 2018: 1 · 2019: 1 · 2020: 1 · 2021: 1 · 2022: 1 · 2023: 1 · 2024: 1 · 2025: 1 · 2026: 1

    AI say-do, 2023 to 2026: 0 AI items announced; 0 GA, 0 still preview, 0 pending, 0 dropped. Red Hat's AI features (Lightspeed, the intelligent assistant, the MCP server) ship in Ansible Automation Platform and are counted there.

    The engine has no AI feature of its own and no official agent interface. Agents can call it only through the command line or the runner library, which is why the AI score sits near zero here while the platform built on it scores.

    Reading

    Ansible is the most widely reached automation engine on this list, and the collections model means its breadth is now set by vendors and community teams as much as by Red Hat. The engine itself has changed slowly since 2020: the real movement is in collections and in the 2.19 templating rework. Its limits are structural: no state file, no drift detection, no control plane and no AI, so every one of those needs a product on top. The thing to check before you build on it is which collection covers your platform and who maintains it, because Proxmox VE and XCP-ng are community or nothing while vSphere, Nutanix and Hyper-V have certified content.

    Open questions and items to re-score

    • Capability steps before 2023 (2.0, 2.9, 2.10) are dated from release pages; their feature content was not read from release notes.
    • Red Hat certification status of community.vmware, community.proxmox and kubevirt.core was not confirmed.
    • The reason for the 2.19 slip from May to July 2025 was not found.
    • Whether an official upstream MCP server exists was not searched exhaustively.
    • Ownership dates (Red Hat acquisition 2015-10, IBM closing 2019-07-09) come from Wikipedia.

    Related reading

    More from Infrastructure Automation Ledger 2026